BJU International (1999), 83, 837–841

A KATP

-channel opener as a potential treatment modality forerectile dysfunctionD.G. MOON, H.S. BYUN and J.J . KIMDepartment of Urology, Korea University Hospital, Seoul, Korea

Objectives To assess the eCects of pinacidil (a KATP

- the cavernosal relaxation induced by acetylcholine,PGE1 or l-arginine (P<0.01). Notably, pinacidilchannel opener) for the treatment of penile erectile

dysfunction and to examine the role of the K+-channel induced cavernosal relaxation after injecting the drugseven in cases refractory to higher concentrations (0.1in cavernosal smooth muscle contractility.

Materials and methods Using a feline model, the magni- mol/L) of the drugs (n=11, P<0.01).Conclusions These results suggest that pinacidil istude of penile erection caused by pinacidil was com-

pared with that caused by erectogenic drugs, e.g. eCective in relaxing feline erectile tissue in vivo,probably via increased K+ permeability and sub-acetylcholine, prostaglandin E1 (PGE1) and l-arginine.

The eCects of K+-channel blockers (4-aminopyridine, sequent hyperpolarization. Further comparative stud-ies with erectogenic compounds on human erectileglibenclamide and tetraethylammonium) and pinacidil

on penile erections induced by the drugs were tissue and clinical testing are required to determinewhether K+-channel openers can be used in theinvestigated.

Results The intra-arterial injection of pinacidil caused a diagnosis and treatment of erectile dysfunction.However, pinacidil seems promising as an intracaver-dose-dependent increase in intracavernosal pressure

(ICP) and the increase in ICP induced by pinacidil nosal agent combined with PGE1 to produce syner-gistic eCects.with acetylcholine, PGE1 or l-arginine was more

pronounced than with the compounds alone. Keywords Erectile dysfunction, K+ channels, pinacidil,therapyFurthermore, pinacidil (1 mmol/L) eCectively reversed

the inhibitory eCects of the K+-channel blockers on

alternatives. Several investigators have shown that pina-Introduction

cidil, a KATP

channel-opener, eCectively relaxes smoothmuscle [7–9]; the present study was designed to investi-Trabecular smooth muscle tone has recently been

accepted as a major contributing factor to trabecular gate the eCect of pinacidil as an alternative componentof intracavernosal injection for the treatment of erectilesmooth muscle contractility; relaxation of the smooth

muscle of the corpus cavernosum is critical in the dysfunction.development of a penile erection [1]. For erection tooccur, the penile arteries and sinusoids must dilate,

Materials and methodsthereby increasing the blood flow into the penis. For thepenis to be flaccid, the smooth muscles of the penile Fifty-five mature male cats (body weight 2.8–4.2 kg)

were anaesthetized with a-chloralose (60 mg/kg intra-arteries and the trabeculae of the corpora cavernosamust be contracted, thereby exerting maximal resistance muscularly); 0.9% normal saline was infused via the

antecubital vein at 2 mL/kg per hour. Ventilation wasagainst arterial flow [2]. Any drug which producesrelaxation when locally injected may also induce ensured via a tracheotomy connected to an animal

ventilator (Harvard Apparatus Ltd, Boston, MA, USA).erection.Current pharmacotherapy for impotence consists of Pancuronium bromide (0.3 mL/kg) was infused as a

neuromuscular blockade; supplemental doses oforal, intracavernosal and topical agents. The drugs mostoften used, e.g. papaverine, phentolamine and PGE1 a-chloralose and pancuronium bromide were given as

needed to maintain a uniform level of anaesthesia. The[3–5], are associated with various side-eCects such aspriapism, corporal fibrosis and pain [6]. Thus, there has internal carotid artery was catheterized centrally to

permit arterial blood gas analysis (ABGA) and to monitorbeen increasing interest in finding eCective and safesystemic arterial blood pressure. A 23 G needle wasplaced into the corpus cavernosum to measure theAccepted for publication 16 December 1998

837© 1999 BJU International

838 D.G. MOON, H.S. BYUN and J.J . KIM

Table 1 Characteristics of the 55 cats usedchanges in intracavernosal pressure (ICP). Both thesystemic arterial blood pressure and ICP were measured

Variable Mean (sd)using a pressure transducer (Gilson T-T23XL, GilsonInst) connected to a physiograph (Gilson IC-MP). A

Body weight (kg) 3.4 (0.45)branch of the internal pudendal artery was microcannul- Body temperature (°C) 35.6 (1.84)ated retrogradely on one side with an elongated poly- Basal ICP (mmHg) 20.7 (6.76)ethylene tube (PE-10 tubing, original inner diameter Arterial blood pressure (mmHg)

Systolic 145.4 (16.35)0.25 mm, outer diameter 0.76 mm) to administer drugsDiastolic 124.7 (13.74)(all in 0.1 mL) into the penis; the remaining branches

Arterial blood gas analysis(except the penile artery) were then ligated. NormoxiapH 7.31 (0.15)(pH > 7.25, PO

2> 100 mmHg, PCO

2< 40 mmHg)

PCO2

(mmHg) 24.7 (5.73)was produced using appropriate ventilator settings (e.g. PO

2(mmHg) 99.4 (13.29)

tidal volume 10 mL/kg; respiratory rate 15–20/min)and monitored using ABGA and a capnometer (Model2200, Traverse Medical Monitors, USA) connected tothe expired air. alone (Fig. 1a), they caused dose-dependent increases in

the ICP (P<0.01). Acetylcholine was the most potentTo assess the eCects of solvents, distilled water, normalsaline, DMSO and alcohol were infused intra-arterially. EC and pinacidil the least. The cavernosal relaxation

induced by 1 mmol/L pinacidil was 30% of that fromIntra-arterial water, saline and DMSO had no eCect onthe ICP, even in large volumes as a bolus. However, 1 mmol/L acetylcholine (n=11, P<0.05). When admin-

istered with the ECs (Fig. 1b) pinacidil augmented theintra-arterial alcohol induced an initial significantincrease in ICP and subsequent unresponsiveness to cavernosal relaxation induced by acetylcholine (by

115%), PGE1 (by 70%) and l-arginine (by 35%) (allexperimental drugs. Possibly alcohol has sclerotic eCectson the intravascular surface. Thus, distilled water was P<0.01).

Pretreatment with 1 mmol/L 4-aminopyridine, TEAused as a solvent for the hydrophilic drugs and DMSOfor hydrophobic solutes. and glibenclamide suppressed acetylcholine-induced cav-

ernosal relaxation, but their eCects were reversed byIn each animal, the basal ICP, DICP (the net changefrom the basal ICP) and changes in systemic blood 1 mmol/L pinacidil (n=11, P<0.01; Table 2). Addition

of 1 mmol/L acetylcholine enhanced the relaxation bypressure were recorded after the intra-arterial injectionof drugs. In initial experiments, control erectile responses 60–90% of the control (P<0.01). The most potent

inhibitor was 4-aminopyridine, but there was no signifi-were established with erectogenic compounds (ECs), i.e.acetylcholine, PGE1 and l-arginine. Responses to pinaci- cant diCerence after the eCects were reversed. Similar

results were obtained with the PGE1-induced cavernosaldil in incremental doses (0.01–100 mmol/L) were thenmonitored. The DICP induced by pinacidil with the ECs relaxation, and again the eCects were reversed by

1 mmol/L pinacidil (n=9, P<0.01; Table 2). PGE1were compared with that for the ECs alone. To determinethe activity of K+-channel blockers, 1 mmol/L of enhanced the relaxation by 60–100% of the control

(P<0.01). Results were also similar for l-arginine-4-aminopyridine, glibenclamide or tetraethylammonium(TEA) were administered and the DICP caused by induced cavernosal relaxation (n=7, P<0.01), with

relaxation enhanced by 30–50% of the control1 mmol/L of the ECs monitored. Pinacidil was thensupplied to determine any reversal of the inhibition by (P<0.05). Again, 4-aminopyridine was the most potent

inhibitor in both cases, with no significant diCerencesK+-channel blockers. All data were presented as themean (sd) and were assessed using Student’s t-test and on reversal.the Mann–Whitney rank-sum test, with P<0.05 con-sidered to indicate statistical significance.

Discussion

Pinacidil is a vasodilator, introduced as an antihyperten-Results

sive agent. The vasodilatory eCect of K+-channel openersis believed to be produced by the opening of K+-channels,The characteristics of the cats are shown in Table 1; the

ABGA values were within the range of normoxia leading to K+ eAux, hyperpolarization and a reductionof Ca2+ influx through the voltage-operated Ca2+ chan-(P<0.01). Acetylcholine and adrenaline induced statisti-

cally significant changes in systemic blood pressure but nels necessary for activating contraction [10,11].Currently available K+-channel openers are exclusivelywere well tolerated by the animals; pinacidil lowered

systemic arterial blood pressure but was relatively well KATP

-channel openers, e.g. nicorandil, cromakalim,pinacidil, minoxidil and diazoxide [12–14].tolerated. When pinacidil and ECs were administered

© 1999 BJU International 83, 837–841

PINACIDIL FOR ERECTILE DYSFUNCTION 839

Fig. 1. Responses of the ICP to theintracavernosal injection of a, no drug(control, dark green) acetylcholine (lightgreen), l-arginine (light red), PGE1 (darkred) and pinacidil alone (black) (n=11), andb, each drug combined with pinacidil. In a,

all compounds caused a dose-dependentincrease in ICP (P<0.05); in b, pinacidilsignificantly augmented the relaxationinduced by the ECs.

Pharmacologically, pinacidil has a short duration of cium sensitivity have also been proposed [16]. Functional(physiological) antagonism of intracellular calciumaction and is less potent than the others [8,15]. Thus

pinacidil was used in the present study as a KATP

-channel levels/myofilament calcium sensitivity is the primarymechanism responsible for the relaxation of smoothopener because of its presumably lesser systemic eCects

than the other compounds, improving the comfort of the muscle. Cyclic AMP and cGMP modulate the continuoustransmembrane calcium flux through the voltage-experimental animals in vivo and inducing fewer side-

eCects if used clinically. Pinacidil reduced systemic dependent calcium channels which are critical to thesustained contraction of corporal smooth musclearterial blood pressure but was well tolerated.

Ultimately, the activation of corporal smooth muscle [17–20]. Increases in intracellular cAMP represents theprimary mechanism by which most intracavernosalcells, either contraction or relaxation, is mediated by the

rise and fall, respectively, of the free intracellular calcium injection agents, e.g. PGE1 and papaverine, mediate theireCects. Increases in intracellular cGMP levels are thoughtconcentration, although changes in myofilament cal-

© 1999 BJU International 83, 837–841

840 D.G. MOON, H.S. BYUN and J.J . KIM

Table 2 ECects of K+-channel blockers and pinacidil (1 mmol/L) on relaxation induced by ECs

Mean (sd) DICP, mmHgDrugapplied 4-AP TEA Glibenclamide

Acetylcholine (n=11)Control 10.2 (3.32) – –+K+-blocker 1.0 (0.95)† 2.3 (1.87)† 3.2 (2.32)†+Pinacidil 5.6 (0.33)† 7.5 (4.63)† 7.9 (0.66)*

PGE1 (n=9)Control 6.4 (2.94) - -+K+-blocker 0.12 (0.0)† 0.8 (0.55)† 1.5 (1.36)†+Pinacidil 2.9 (2.34)* 2.0 (1.36) 4.0 (3.28)*

l-arginine (n=7)Control 7.8 (4.49) – –+K+-blocker 0.09 (0.0)† 2.0 (1.56)† 2.5 (1.49)+Pinacidil 2.7 (1.32)* 3.0 (1.85) 3.6 (2.19)†

*P<0.05; †P<0.01 vs corresponding values for K+-blockers. 4-AP, 4-aminopyridine; TEA, tetraethylammonium.

to be the primary mechanism by which nitric oxide (NO) is the maxi-K channel. In the present study, theEC-induced relaxations were suppressed by TEA, gliben-released from nonadrenergic noncholinergic (NANC)

nerves exerts its relaxant eCect, after the activation of clamide or 4-aminopyridine. This may be because apartfrom activating the maxi-K channel, PKA or PKC notsoluble guanylate cyclase.

The K+ ion flux is controlled by three main eCector only open the KATP

channel but also suppress the calciumchannel. These results concur with those of Satake et al.pathways [21]; the first two are the cAMP/protein kinase

A (PKA) and cGMP/protein kinase G (PKG) pathways [23], who concluded that the maxi-K, KATP

and voltage-dependent (K

v) channels are involved in vasorelaxantwhich are activated by, e.g. PGE1 and NO, respectively.

While these pathways apparently modulate the activity responses to acetylcholine in rat aortic rings.The K

ATPchannel is metabolically regulated andof the maxi-K channel, their eCects on the K

ATP-channel

have not been documented, but seem likely. A third another important modulator of corporal smooth muscletone [24–27]. Glibenclamide is a K

ATPchannel-specificpathway is by K+-channel modulators, which exert their

relaxant actions on corporal smooth muscle, at least blocker, 4-aminopyridine a nonspecific Kv-channel

blocker and TEA a specific Kv

blocker but with additionalpartly by altering the activity of the KATP

channel. Theincrease in ICP induced by pinacidil with the ECs was action on the maxi-K and K

ATPchannel. Pinacidil eCec-

tively reversed the inhibitory eCects of the K+-channelmore pronounced than with the ECs alone; pinacidilinduced an additional 35–115% of relaxation when blockers on cavernosal relaxation induced by the ECs.

An ATP-sensitive channel, which can be reversed bygiven with the ECs. This may arise by the dual activationof the K

ATPchannel and maxi-K channel. Notably, pinacidil, is probably involved, but the participation of

other channel types cannot be excluded. The mechan-pinacidil induced cavernosal relaxation with ECs even incases refractory to higher concentrations (0.1 mmol/L) isms of action of 4-aminopyridine should be elucidated.

The results of the ABGA showed normoxia, but unlikeof ECs alone. These results suggest that the KATP

-channelopener may be useful in patients who cannot obtain an the in vitro studies, a problem in vivo is sympathetic

hyperactivity. Apart from other nonrelaxing factors,erection with other intracavernosal agents, especiallyPGE1. The synergistic eCects of pinacidil with PGE1 have neurogenic contraction should be considered as a poss-

ible bias in the in vivo pharmacological studies. In thepotential therapeutic implications for their use as acombination. present study, intra-arterial infusion was used to deliver

the drugs rather than direct cavernosal injection.The maxi-K channel (KCa

) subtype is similar to thecalcium-sensitive K channel and is clearly the predomi- Cavernosal smooth muscle relaxation induced by NO

donors causes erection (via an increase in ICP) andnant K+-channel subtype found in corporal smoothmuscle. It is responsible for #90% of the observed penile lengthening [28]; in the present study, the ICP

was recorded rather than penile length.whole-cell outward K current seen in cultured humancorporal smooth muscle cells [22]. It is believed that the In conclusion, the present results suggest that the

KATP

and Kv

channel, including the maxi-K channel,main pathway of acetylcholine, PGE1 or l-arginine entry

© 1999 BJU International 83, 837–841

PINACIDIL FOR ERECTILE DYSFUNCTION 841

15 Holmquist F, Andersson KE, Fovaeus M, Hedlund H. K+-play major roles in penile erection under normoxia.channel openers for relaxation of isolated penile erectilePinacidil has potential as an erectogenic agent and maytissues from rabbit. J Urol 1990; 144: 146–51be clinically useful as an intracavernosal agent in combi-

16 Somlyo AP, Somlyo AV. Signal transduction and regulationnation with PGE1. However, further studies of KATP

-in smooth muscle. Nature 1994; 372: 231–6channel openers and the mechanisms of K+-channels

17 Lincoln TM, Cornwell TL, Taylor AE. cGMP-dependentare required if such agents are to be used in the diagnosis

protein kinase mediates the reduction of Ca2+ by cAMP inand treatment of erectile dysfunction. vascular smooth muscle cells. Am J Physiol 1990; 258:

C399–40718 Lincoln TM, Cornwell TL. Towards an understanding of

the mechanism of action of cyclic AMP and cyclic GMP inReferences smooth muscle relaxation. Blood Vessels 1991; 28: 129–37

1 Andersson KE, Wagner G. Physiology of penile erection. 19 Cornwell TL, Arnold E, Boerth NJ, Lincoln TM. InhibitionPhysiol Rev 1995; 75: 191–236 of smooth muscle cell growth by nitric oxide and activation

2 Lue TF, Tanagho EA. Physiology of erection and pharmaco- of cAMP-dependent protein kinase by cGMP. Am J Physiollogical management of impotence. J Urol 1987; 137: 1405–13; 1994; 267: C829–36 20 Jiang H, Colbran JL, Francis SH, Corbin JD. Direct evidence

3 Juenemann KP, Lue TF, Fournier GR Jr, Tanagho EA. for cross activation of cGMP-dependent protein kinase byHemodynamics of papaverine and phentolamine-induced cAMP in pig coronary arteries. J Biol Chem 1991;penile erection. J Urol 1986; 136: 158–61 267: 1015–9

4 Govier FE, McClure RD, Weissman RM, Gibbons RP, 21 Christ GJ, Richards S, Winkler A. Integrative erectilePritchett TR, Kramer-Levin K. Experience with triple-drug biology: the role of signal transduction and cell-to-celltherapy in pharmacological erection program. J Urol 1993; communication in coordinating corporal smooth muscle150: 1822–4 tone and penile erection. Int J Impot Res 1997; 9: 69–84

5 Gerber SC, Levine LA. Pharmacological erection program 22 Christ GJ, Spray DC, Brink PR. Characterization of Kusing prostaglandin E1. J Urol 1991; 146: 786–91 current in cultured human corporal smooth muscle cells.

6 Althof SE, Turner LA, Levin SB et al. Why do so many J Androl 1993; 14: 319–28people drop out from auto-injection therapy for impotence? 23 Satake N, Shibata M, Shibata S. The involvement of KCa,J Sex Marital Ther 1989; 15: 121–9 KATP and KV channels in vasorelaxing responses to

7 Hellstrom WJG, Wang R, Kadowitz PJ, Domer FR. Potassium acetylcholine in rat aortic rings. Gen Phamacol 1997;channel agonists cause penile erection in cats. Int J Impot 28: 453–7Res 1992; 4: 35–43 24 Rudy B. Diversity and ubiquity of K channels. Neuroscience

8 Holmquist F, Andersson KE, Hedlund H. ECects of pinacidil 1988; 25: 729–49on isolated human corpus cavernosum penis. Acta Physiol 25 Ibbotson T, Edwards G, Noack T, Weston AH. ECects ofScand 1990; 138: 463–9 P1060 and apikalim on whole-cell currents in rat portal

9 Giraldi A, Wagner G. ECects of pinacidil upon penile erectile vein; inhibition by glibenclamide and phentolamine. Brtissue, in vitro and in vivo. Pharmacol Toxicol 1990; J Pharmacol 1993; 108: 991–867: 235–8 26 Stanfield PR. Tetraethylammonium ions and the potassium

10 Videbaek LM, Aalkjaer C, Mulvany MJ. ECect of pinacidil permeability of excitable cells. Rev Physiol Biochemon norepinephrine- and potassium-induced contractions Pharmacol 1983; 97: 1–67and membrane potential in rat and human resistance 27 Strong PN. Potassium channel toxins. Pharmacol Thervessels and in rat aorta. J Cardiovasc Pharmacol 1988; 1990; 46: 137–6212: S23–9 28 Wang R, Domer FR, Sikka SS, Kadowitz PJ, Hellstrom WJG.

11 Hermsmeyer RK. Pinacidil actions on ion channels in Nitric oxide mediates penile erection in cats. J Urol 1994;vascular muscle. J Cardiovasc Pharmacol 1988; 12: S17–22 151: 234–7

12 Andersson KE. Clinical pharmacology of potassium channelopeners. Pharmacol Toxicol 1992; 70: 244–54

Authors13 Newgreen DT, Bray KM, McHarg AD et al. The action ofdiazoxide and minoxidil sulfate on rat blood vessels: a D.G. Moon, MD, Clinical & Research Fellow in Urology.

H.S. Byun, MD, Clinical Fellow in Urology.comparison with cromakalim. Br J Pharmacol 1990;100: 605–13 J.J. Kim, MD, Professor in Urology.

Correspondence: Dr J.J. Kim, Department of Urology, Korea14 Edwards G, Weston AH. The pharmacology of ATP-sensitivepotassium channels. Annu Rev Pharmacol Toxicol 1993; 33: University Hospital, 126–1, 5-ka, Anam-dong, Sungbuk-ku,

Seoul 136–705, Korea.597–637

© 1999 BJU International 83, 837–841