European Journal of Pharmacology, 215 (1992) 177-183
177
(~ 1992 Elsevier Science Publishers B.V. All rights reserved 0014-2999/92/$05.00
EJP 52436
Effects of K + channel blockers on the relaxant action of dihydralazine, c r o m a k a l i m and nitroprusside in isolated rabbit femoral arteries Steffen Thirstrup and Jens Erik Nielsen-Kudsk Institute ~f Pharmacology3', Tile Bartholin Building~ Uni~'ersity of Aarhus, DK-8000 Aarhus C, Denmark Received 14 January 1992, accepted 18 February 1992
The relaxant responses to dihydralazine and the influence of different K + channel blockers were studied in isolated rabbit femoral arteries. The prototype K + channel opener, cromakalim, and nitroprusside, which does not produce relaxation by K + channel activation were used for comparison. Dihydralazine was most effective on contractions induced by noradrenaline (EC50 = 1.1 /xM; E ..... = 95%) and relaxed the contractions elicited by 20 mM K + (ECs~) = 2.0 #M; Em~,x= 81% in preference to 124 mM K+-induced contractions (ECs0 = 30.1 /xM; Ema x = 5 4 % ) . Cromakalim, but not nitroprusside, also selectively relaxed 20 mM K+-induced contractions. In noradrenaline-contracted arteries, glibenclamide (10 p.M) completely suppressed the relaxant response to cromakalim but did not influence the vasorelaxation produced by dihydralazine or nitroprusside. Tetraethylammonium (8 mM) and Cs + (4 mM) shifted the concentration-relaxation curve for dihydralazine 2-fold to the right, whereas Ba 2+ (0.1 raM), 4-aminopyridine (5 mM) and procaine (0.1 mM) failed to influence dihydralazine-induced responses. Tctracthylammonium (8 raM) shifted the concentration-relaxation curve for cromakalim and nitroprusside 6-fold to the right and suppressed the maximal relaxant effects by about 30%. It is concluded that dihydralazinc produces vascular smooth muscle relaxation by a mechanism different from the oper~ing of glibenclamide- and ATP-sensitive K + channels. Dihydralazine; Hydralazine; Cromakalim; Nitroprusside; K + channels; Smooth muscle (vascular)
1. Introduction
Hydralazine (1-hydrazinophthalazine) is a classical vasodilator drug which decreases systemic vascular resistance by direct relaxation of the smooth muscle of arteries and arterioles. It has been used in the treatment of hypertension for decades and in recent years for the treatment of congestive heart failure also. Dihydralazine, which contains a hydrazine group in both the 1- and 4-position of the phthalazine molecule, has in vitro and in vivo vasodilator actions that are indistinguishable from those p r o d u c e d by h y d r a l a z i n e (Haeusler and Gerold, 1979). Only dihydralazine is available for parenteral use in most E u r o p e a n countries, where it is used in hypertensive emergencies and in the t r e a t m e n t of preeclampsia. A l t h o u g h n u m e r o u s theories have been put forward (Kreye, 1984), the mechanism of action of the hydralazines is still not clear. These drugs do not act like a d r e n o c e p t o r antagonists (Khayyal et al., 1981), cal-
Correspondence to: J.E. Nielsen-Kudsk, Institute of Pharmacology, The Bartholin Building, University of Aarhus, DK-8000 Aarhus C, Denmark. Tel. 45.86.202711 ext. 3761, fax 45.86.191 277.
cium antagonists ( H e r m s m e y e r et al., 1983; Orallo et al., 1991) or enhancers of c - A M P or c - G M P production (Kauffman et al., 1986). M e m b r a n e hyperpolarization due to activation of K + channels has recently been recognized as an important m o d e of action for several vasodilators (Standen et at., 1989; Brayden et al., 1991). Cromakalim and pinacidil are prototype drugs for this rapidly increasing g r o u p of smooth muscle relaxants classified as K + channel o p e n e r s (Hamilton and Weston, 1989). T h e antihypertensive agents, minoxidil sulphate and diazoxide, which previously were g r o u p e d together with hydralazine as direct vasodilators with an u n k n o w n mechanism of action, have also turned out to be K + channel o p e n e r s (Meisheri et al., 1988; Newgreen at al., 1990). It has been d e m o n s t r a t e d that hydralazine hyperpolarizes p h e n y l e p h r i n e - c o n t r a c t e d rat caudal artery ( H e r m s m e y e r et al., 1983) and that it is without relaxant activity on the isolated rat perfused mesenteric vascular bed or the isolated rabbit renal artery when contractions are induced by high concentrations of K + (Ally and Horrobin, 1986; Khayyal et al., 1981). Such effects are compatible with K + channel activation. T h e present study was carried out to determine if dihydralazine produces vasorelaxation characteristic of
178
K ÷ channel opening. The effects of the drug on contractions induced by noradrenaline (NA), 20 mM K + or 124 mM K +, in the isolated rabbit femoral artery were investigated. Six K + channel blockers (glibenclamide, tetraethylammonium (TEA), 4-aminopyridine, procaine, BaC12, CsC1) were tested for a possible inhibitory action on the dihydralazine responses. The K + channel opener, cromakalim, and another vasodilator, nitroprusside, which does not produce relaxation by opening of K + channels, were studied for comparison with dihydralazine.
2. Materials and methods
2.1. Arterial preparations and measurement of contractile force Albino rabbits of either sex (body weight 2589 _+ 105 S.E., n = 22) were given heparin (75 1 U / k g i.v.) and anesthetized with thiopental (50 m g / k g i.v.). The femoral arteries were rapidly removed from the animals and tubular segments with a length of 2 mm were prepared under the microscope. Six arterial preparations were taken from the same animal at each experimental session and were transferred to six 5-ml organ baths containing Krebs solution (composition in raM: NaCI 118.0, KC1 4.6, CaCl 2 2.5, MgSO 4 1.15, N a H C O 3 24.9, K H 2 P O 4 1.15, glucose 5.5). The bath had thermostats set at 3 7 ° C and were continuously aerated with a mixture of 95% 0 2 and 5% CO 2 (pH = 7.4). For recording of isometric contractile force, each preparation was mounted in a myograph between two fine stainless steel pins submerged in each bath. The myograph construction was described in detail earlier (Nielsen-Kudsk et al., 1986). Signals from the straingauge transducers were displayed on a six-channel Watanabe linearcorder (WR 3101). The arterial preparations were initially suspended under a passive force of 0.8 g which was optimal for contractile responses to 124 mM K +. The preparations were allowed to equilibrate for 45 rain before addition of any drugs. All experiments were then initiated with a 124 mM K + contraction to ensure adequate reactivity of each arterial segment.
2.2. Experiments The preparations were contracted with NA 1 # M , 20 mM isoosmolar K + Krebs solution or 124 mM isoosmolar K + Krebs solution. When a stable contraction level was attained, dihydralazine was added cumulatively to the organ baths. Concentration increases were made every 15 min when the dihydralazine responses were apparently stable. Concentration-relaxation curves for dihydralazine were obtained in NA-con-
tracted arteries in the absence and in the presence of different K + channel blockers: glibenclamide (10/xM), T E A (8 raM), 4-aminopyridine (5 mM), procaine (0.1 raM), BaC1, (0.1 raM) or CsCI (4 raM). Each K + channel blocker was added when a stable NA contraction had developed. All K + channel blockers caused temporary changes in NA-induced tone that were allowed to disappear before dihydralazine addition (after 15-30 rain). Only a single concentration-relaxation curves to dihydralazine could be obtained with each preparation. Despite thorough washing of the preparations, the responses to the contractile agents could not be restored fully after cumulative dihydralazine addition, indicate a rather firm tissue binding of the drug. Ascorbic acid (1 raM) was present in the baths to avoid oxidative degradation of the drugs. All experiments were carried out under sodium light in order to avoid possible photolysis of dihydralazine. The relaxant effect of dihydralazine on NA-induced contractions was also tested in preparations without endothelium. The endothelium was removed from the tubular arterial preparations by rubbing the inside of the vessels with a hair from a horse tail. The effectiveness of this procedure was verified by the disappearance of the acetylcholine vasodilator response. The effect of methylene blue (50 /xM) on dihydralazine responses was also tested. Concentration-effect curves for cromakalim and nitroprusside were obtained as described for dihydralazine, but only selected K + channel blockers, glibenclamide (10 ~M) and T E A (8 raM), were tested for effects on relaxation produced by these drugs. A concentration-relaxation curve for hydralazine was obtained in NA-contracted arteries for comparison with dihydralazine. Only one concentration-effect curve was obtained for each of the drugs in each arterial preparation.
2.3. Data analysis The results are given as means _+ S.E. Relaxant responses to the vasodilators were measured as percentage suppression of contractile force. Concentration-relaxation data were fitted to the Hill equation, E = E .... C~/(EC;0 + CO, by iterative non-linear regression analysis. The Hill coefficient (S) expresses the slope of the concentration-relaxation curve. The two-tailed Student's t-test for unpaired comparisons was used for statistical evaluation. A P < 0.05 was considered significant.
2.4. Drugs The drugs used were dihydralazine (Nepresol
177
(~ 1992 Elsevier Science Publishers B.V. All rights reserved 0014-2999/92/$05.00
EJP 52436
Effects of K + channel blockers on the relaxant action of dihydralazine, c r o m a k a l i m and nitroprusside in isolated rabbit femoral arteries Steffen Thirstrup and Jens Erik Nielsen-Kudsk Institute ~f Pharmacology3', Tile Bartholin Building~ Uni~'ersity of Aarhus, DK-8000 Aarhus C, Denmark Received 14 January 1992, accepted 18 February 1992
The relaxant responses to dihydralazine and the influence of different K + channel blockers were studied in isolated rabbit femoral arteries. The prototype K + channel opener, cromakalim, and nitroprusside, which does not produce relaxation by K + channel activation were used for comparison. Dihydralazine was most effective on contractions induced by noradrenaline (EC50 = 1.1 /xM; E ..... = 95%) and relaxed the contractions elicited by 20 mM K + (ECs~) = 2.0 #M; Em~,x= 81% in preference to 124 mM K+-induced contractions (ECs0 = 30.1 /xM; Ema x = 5 4 % ) . Cromakalim, but not nitroprusside, also selectively relaxed 20 mM K+-induced contractions. In noradrenaline-contracted arteries, glibenclamide (10 p.M) completely suppressed the relaxant response to cromakalim but did not influence the vasorelaxation produced by dihydralazine or nitroprusside. Tetraethylammonium (8 mM) and Cs + (4 mM) shifted the concentration-relaxation curve for dihydralazine 2-fold to the right, whereas Ba 2+ (0.1 raM), 4-aminopyridine (5 mM) and procaine (0.1 mM) failed to influence dihydralazine-induced responses. Tctracthylammonium (8 raM) shifted the concentration-relaxation curve for cromakalim and nitroprusside 6-fold to the right and suppressed the maximal relaxant effects by about 30%. It is concluded that dihydralazinc produces vascular smooth muscle relaxation by a mechanism different from the oper~ing of glibenclamide- and ATP-sensitive K + channels. Dihydralazine; Hydralazine; Cromakalim; Nitroprusside; K + channels; Smooth muscle (vascular)
1. Introduction
Hydralazine (1-hydrazinophthalazine) is a classical vasodilator drug which decreases systemic vascular resistance by direct relaxation of the smooth muscle of arteries and arterioles. It has been used in the treatment of hypertension for decades and in recent years for the treatment of congestive heart failure also. Dihydralazine, which contains a hydrazine group in both the 1- and 4-position of the phthalazine molecule, has in vitro and in vivo vasodilator actions that are indistinguishable from those p r o d u c e d by h y d r a l a z i n e (Haeusler and Gerold, 1979). Only dihydralazine is available for parenteral use in most E u r o p e a n countries, where it is used in hypertensive emergencies and in the t r e a t m e n t of preeclampsia. A l t h o u g h n u m e r o u s theories have been put forward (Kreye, 1984), the mechanism of action of the hydralazines is still not clear. These drugs do not act like a d r e n o c e p t o r antagonists (Khayyal et al., 1981), cal-
Correspondence to: J.E. Nielsen-Kudsk, Institute of Pharmacology, The Bartholin Building, University of Aarhus, DK-8000 Aarhus C, Denmark. Tel. 45.86.202711 ext. 3761, fax 45.86.191 277.
cium antagonists ( H e r m s m e y e r et al., 1983; Orallo et al., 1991) or enhancers of c - A M P or c - G M P production (Kauffman et al., 1986). M e m b r a n e hyperpolarization due to activation of K + channels has recently been recognized as an important m o d e of action for several vasodilators (Standen et at., 1989; Brayden et al., 1991). Cromakalim and pinacidil are prototype drugs for this rapidly increasing g r o u p of smooth muscle relaxants classified as K + channel o p e n e r s (Hamilton and Weston, 1989). T h e antihypertensive agents, minoxidil sulphate and diazoxide, which previously were g r o u p e d together with hydralazine as direct vasodilators with an u n k n o w n mechanism of action, have also turned out to be K + channel o p e n e r s (Meisheri et al., 1988; Newgreen at al., 1990). It has been d e m o n s t r a t e d that hydralazine hyperpolarizes p h e n y l e p h r i n e - c o n t r a c t e d rat caudal artery ( H e r m s m e y e r et al., 1983) and that it is without relaxant activity on the isolated rat perfused mesenteric vascular bed or the isolated rabbit renal artery when contractions are induced by high concentrations of K + (Ally and Horrobin, 1986; Khayyal et al., 1981). Such effects are compatible with K + channel activation. T h e present study was carried out to determine if dihydralazine produces vasorelaxation characteristic of
178
K ÷ channel opening. The effects of the drug on contractions induced by noradrenaline (NA), 20 mM K + or 124 mM K +, in the isolated rabbit femoral artery were investigated. Six K + channel blockers (glibenclamide, tetraethylammonium (TEA), 4-aminopyridine, procaine, BaC12, CsC1) were tested for a possible inhibitory action on the dihydralazine responses. The K + channel opener, cromakalim, and another vasodilator, nitroprusside, which does not produce relaxation by opening of K + channels, were studied for comparison with dihydralazine.
2. Materials and methods
2.1. Arterial preparations and measurement of contractile force Albino rabbits of either sex (body weight 2589 _+ 105 S.E., n = 22) were given heparin (75 1 U / k g i.v.) and anesthetized with thiopental (50 m g / k g i.v.). The femoral arteries were rapidly removed from the animals and tubular segments with a length of 2 mm were prepared under the microscope. Six arterial preparations were taken from the same animal at each experimental session and were transferred to six 5-ml organ baths containing Krebs solution (composition in raM: NaCI 118.0, KC1 4.6, CaCl 2 2.5, MgSO 4 1.15, N a H C O 3 24.9, K H 2 P O 4 1.15, glucose 5.5). The bath had thermostats set at 3 7 ° C and were continuously aerated with a mixture of 95% 0 2 and 5% CO 2 (pH = 7.4). For recording of isometric contractile force, each preparation was mounted in a myograph between two fine stainless steel pins submerged in each bath. The myograph construction was described in detail earlier (Nielsen-Kudsk et al., 1986). Signals from the straingauge transducers were displayed on a six-channel Watanabe linearcorder (WR 3101). The arterial preparations were initially suspended under a passive force of 0.8 g which was optimal for contractile responses to 124 mM K +. The preparations were allowed to equilibrate for 45 rain before addition of any drugs. All experiments were then initiated with a 124 mM K + contraction to ensure adequate reactivity of each arterial segment.
2.2. Experiments The preparations were contracted with NA 1 # M , 20 mM isoosmolar K + Krebs solution or 124 mM isoosmolar K + Krebs solution. When a stable contraction level was attained, dihydralazine was added cumulatively to the organ baths. Concentration increases were made every 15 min when the dihydralazine responses were apparently stable. Concentration-relaxation curves for dihydralazine were obtained in NA-con-
tracted arteries in the absence and in the presence of different K + channel blockers: glibenclamide (10/xM), T E A (8 raM), 4-aminopyridine (5 mM), procaine (0.1 raM), BaC1, (0.1 raM) or CsCI (4 raM). Each K + channel blocker was added when a stable NA contraction had developed. All K + channel blockers caused temporary changes in NA-induced tone that were allowed to disappear before dihydralazine addition (after 15-30 rain). Only a single concentration-relaxation curves to dihydralazine could be obtained with each preparation. Despite thorough washing of the preparations, the responses to the contractile agents could not be restored fully after cumulative dihydralazine addition, indicate a rather firm tissue binding of the drug. Ascorbic acid (1 raM) was present in the baths to avoid oxidative degradation of the drugs. All experiments were carried out under sodium light in order to avoid possible photolysis of dihydralazine. The relaxant effect of dihydralazine on NA-induced contractions was also tested in preparations without endothelium. The endothelium was removed from the tubular arterial preparations by rubbing the inside of the vessels with a hair from a horse tail. The effectiveness of this procedure was verified by the disappearance of the acetylcholine vasodilator response. The effect of methylene blue (50 /xM) on dihydralazine responses was also tested. Concentration-effect curves for cromakalim and nitroprusside were obtained as described for dihydralazine, but only selected K + channel blockers, glibenclamide (10 ~M) and T E A (8 raM), were tested for effects on relaxation produced by these drugs. A concentration-relaxation curve for hydralazine was obtained in NA-contracted arteries for comparison with dihydralazine. Only one concentration-effect curve was obtained for each of the drugs in each arterial preparation.
2.3. Data analysis The results are given as means _+ S.E. Relaxant responses to the vasodilators were measured as percentage suppression of contractile force. Concentration-relaxation data were fitted to the Hill equation, E = E .... C~/(EC;0 + CO, by iterative non-linear regression analysis. The Hill coefficient (S) expresses the slope of the concentration-relaxation curve. The two-tailed Student's t-test for unpaired comparisons was used for statistical evaluation. A P < 0.05 was considered significant.
2.4. Drugs The drugs used were dihydralazine (Nepresol
