Vasorelaxant effects of the potassium channel opener SR 47063 on the isolated human saphenous vein and rat aorta

The vasorelaxant effects of SR 47063 (4-(2-cyanimino-1,2-dihydropyrid-1-yl)-2,2-dimethyl-6-nitrochromene), a new K+-channel opener structurally related to levcromakalim, were examined in isolated human saphenous vein (HSV) and rat aorta (RA). HSV or RA rings were precontracted with either KCl or noradrenaline and cumulative relaxant concentration-response curves were obtained for SR 47063 (0.1 nM to 1 μM) in the presence or absence of 3 μM glibenclamide. SR 47063 potently relaxed HSV and RA precontracted with 20 mM (but not 60 mM) KCl or 10 μM noradrenaline in a concentrationdependent manner, showing slightly greater activity in the aorta. The potency of the effect of SR 47063 on HSV and RA was 12and 58-fold greater, respectively, than that reported for the structurally related K+channel opener levcromakalim. The vasorelaxant action of SR 47063 in both blood vessels was strongly inhibited by 3 μM glibenclamide, consistent with a mechanism of action involving ATP-dependent K+channels. Correspondence


Introduction
SR 47063 (4-(2-cyanimino-1,2-dihydropyrid-1-yl)-2,2-dimethyl-6-nitrochromene) is a potassium channel opener (K + -channel opener) currently in clinical trials that has been shown to exert hypotensive effects on rats in vivo (1) and protective effects on ventricular function in cardioplegia (2). Several studies have demonstrated that SR 47063 exhibits a profile of action consistent with that of a K + -channel opener in rat vascular (3,4) and human airway smooth muscle (5). Current electrophysiological studies suggest that this structurally diverse group of drugs may act via the opening of relatively small conductance, glibenclamide-sensitive K +channels in both arterial and venous smooth muscle (6)(7)(8)(9).
Existing data suggest that SR 47063 is more potent than structurally related drugs such as cromakalim and bimakalim in guinea pig heart (3,10). However, comparatively little data exist regarding the effects of SR 47063 on isolated blood vessels, with no published reports about the human vasculature. We have previously described the actions of levcromakalim and P1060, a struc-turally unrelated K + -channel opener, on the isolated human saphenous vein (11). Therefore we decided to assess the potency of SR 47063 in vascular tissue using the isolated human saphenous vein and the rat aorta, and to examine whether the relaxant action of this drug was consistent with that of a K +channel opener. Preliminary data from this study have previously been presented to the British Pharmacological Society (12).

Preparation of blood vessels
Segments of branches of the long saphenous vein (leftovers) were obtained from patients undergoing heart revascularization surgery at the Pedro Ernesto University Hospital, Universidade do Estado do Rio de Janeiro (UERJ). Institutional approval for use of this tissue was obtained. On removal from the patient, the vessel segments were immediately placed in Krebs-Henseleit solution of the following composition: 118.3 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl 2 , 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , 25 mM NaHCO 3 , 0.026 mM EDTA, and 11.1 mM glucose, carefully cleaned of perivascular tissue and cut into approximately 0.3-cm long rings. All experiments were performed on the day of surgery.
Male Wistar rats (250-350 g) were killed by stunning and cervical dislocation. The thoracic aorta was removed by careful dissection, placed in Krebs-Henseleit solution and cut into approximately 0.3-cm long rings.

Organ chamber experiments
Rings of saphenous vein and aorta were suspended in organ chambers filled with 30 ml of Krebs-Henseleit solution bubbled with 95% O 2 and 5% CO 2 at 37 o C. Each ring was suspended by two stainless steel stirrups passed through its lumen. One stirrup was anchored inside the organ chamber and the other was connected to a force transducer (FTA 10; Hewlett-Packard Co., Palo Alto, CA, USA) for the measurement of isometric force with a Hewlett-Packard 7754A recorder. All rings were progressively stretched to the optimal point of the length-tension curve as determined by the response to 60 mM KCl (approximately 4 applications of KCl) and allowed to equilibrate for 60 min.

Experimental protocols
Following equilibration, blood vessel rings were exposed to either 3 µM glibenclamide or appropriate vehicle for 15 min before spasmogen addition. Either KCl (20 mM or 60 mM) or noradrenaline (10 µM) was used to precontract the tissues. Following the development of a stable plateau contraction in response to the spasmogen, SR 47063 (0.1 nM to 1 µM) was applied cumulatively to the organ bath in log increments and concentration-response curves were constructed.

Drugs
SR 47063 (Sanofi Recherche, Montpellier, France) and noradrenaline (Sigma Chemical Co., St. Louis, MO, USA) were prepared as stock solutions (3 and 10 mM, respectively) in absolute ethanol and diluted in distilled water on the day of the experiment. Glibenclamide (Sigma) was prepared as a stock solution (10 mM) in dimethyl sulfoxide and diluted in distilled water.

Statistical analysis
Data are reported as the mean ± SEM of N observations. IC 50 values were calculated for each experimental concentration-effect curve and expressed as the mean of N experiments together with the appropriate SEM. Data were analyzed statistically by the unpaired Student t-test.

Relaxant effects of SR 47063 on KCl-and noradrenaline-induced contraction of the human saphenous vein
KCl (20 mM) induced contraction of the human saphenous vein with a plateau response of 1.34 ± 0.10 g (N = 20, 9 patients). Cumulative application of SR 47063 (0.1 nM to 1 µM) elicited a concentration-dependent relaxation of KCl-induced tone, an effect that was maximal at 0.1 µM ( Figure  1A). However, in the presence of 3 µM glibenclamide the relaxant concentration-response curve was shifted to the right by approximately 50-fold (Table 1; Figure 1A). In saphenous vein rings precontracted with 60 mM KCl, SR 47063 at concentrations up to 1 µM did not induce relaxation (N = 5). Noradrenaline (10 µM) elicited a greater contraction of the saphenous vein than 20 mM KCl, with a stable plateau response of 2.34 ± 0.32 g (N = 14, 6 patients). SR 47063 induced a concentration-dependent inhibition of this developed tone with a reduction of 87% at a concentration of 1 µM ( Figure  1B). SR 47063 was slightly less potent against noradrenaline-induced compared with KClinduced contraction ( Table 1). The relaxation of noradrenaline-precontracted veins was also inhibited by glibenclamide; at the highest concentration of SR 47063 tested (1 µM) approximately 75% of the contraction  induced by noradrenaline remained in the presence of glibenclamide and thus precluded calculation of the IC 50 value.

Relaxant effects of SR 47063 on KCl-and noradrenaline-induced contraction of the rat aorta
Application of 20 mM KCl to the aorta induced a biphasic contraction with a sustained plateau response of 0.81 ± 0.12 g (N = 16). SR 47063 relaxed the precontracted aorta in concentration-dependent manner, with a complete reduction of induced tone at a concentration of 1 µM (Figure 2A). This action was inhibited significantly in the presence of glibenclamide (3 µM) with an approximate 57-fold shift of the IC 50 value (Figure 2A; Table 1). Noradrenaline (10 µM) induced a plateau contraction of 0.91 ± 0.13 g (N = 14) which was fully relaxed by cumulative application of SR 47063 ( Figure 2B).
In the presence of 3 µM glibenclamide this effect was greatly inhibited, with approximately 56% of the contraction remaining at a concentration of 1 µM SR 47063.

Discussion
The present study is the first description of the effects of SR 47063 on isolated human vasculature, extending previous pre-clinical studies performed on animal tissues. We found that SR 47063 has a potent relaxant action on 20 mM KCl-and noradrenalineprecontracted human saphenous vein. However, this relaxant effect was not observed in the presence of a high concentration of potassium (60 mM). Thus, SR 47063 exhibits a profile of pharmacological activity characteristic of K + -channel openers in a variety of vascular tissues including rabbit aorta, dog coronary and middle cerebral artery (13)(14)(15), and the current results are in agreement with previous studies with this drug on rat vascular (3,4), human bronchial (5) and frog skeletal muscle preparations (16,17).
The relaxant effects of SR 47063 on both saphenous vein and aorta were very sensitive to inhibition by glibenclamide, an inhibitor of ATP-dependent K + -channels (K ATP ) in vascular smooth muscle (7), consistent with an action of SR 47063 on this type of K + -channel. A recent electrophysiological investigation has shown that submicromolar concentrations of SR 47063 potently activate glibenclamide-sensitive K + -currents in guinea pig cardiac muscle cells (18), whilst higher concentrations also activate ATP-sensitive channels in skeletal muscle (16). However, we are unaware of any published data regarding the single-cell electrophysiology of SR 47063 in vascular smooth muscle, although the bulk of evidence appears to suggest that SR 47063, in common with its structural analogue levcromakalim, exerts its vascular relaxant effects via the opening of K ATP .
In the guinea pig heart SR 47063 is 25-30-fold more potent in decreasing coronary perfusion pressure than the structurally related benzopyran cromakalim (3). We have previously reported that levcromakalim relaxes 20 mM KCl-induced contractions of the human saphenous vein with an IC 50 value of 97 nM (10) and thus our present results indicate that SR 47063 is approximately 12fold more active than levcromakalim in this tissue. This is in agreement with previous data indicating that SR 47063 is approximately 14-fold more potent than levcromakalim in inhibiting the spontaneous contractile activity of another blood vessel, the rat portal vein, with respective IC 50 values of 5.6 nM (3) and 80 nM (19). Interestingly, comparison of the present study with a re- port on the rat aorta (20) also indicates that SR 47063 is approximately 58-fold more potent against 20 mM KCl-induced contractions than its structural analogue levcromakalim. Indeed this potency difference extends to in vivo decreases in mean arterial blood pressure, with SR 47063 being about 6-fold and 15-fold more potent than cromakalim in spontaneously hypertensive and normotensive anesthetized rats, respectively (1). At present no published comparative clinical data on humans are available to confirm or refute this putative potency advantage of SR 47063 over levcromakalim.
In conclusion, we have shown that SR 47063 possesses a potent vasorelaxant activity on both human and animal blood vessels, exhibiting effects consistent with the opening of K ATP channels. This compound appears to possess a degree of selectivity for vascular smooth muscle since higher concentrations of SR 47063 are required to elicit comparable relaxations of human bronchial (5) and bladder smooth muscle (21), and guinea pig cardiac muscle (10). Thus, our present results support the possibility of a future therapeutic use for SR 47063 in the treatment of hypertension.