Heart 1996;76:109-1 16

REVIEW

K+ channel opening: a new drug principle incardiovascular medicine

Jens Erik Nielsen-Kudsk, S0ren Boesgaard, Jan Aldershvile

Many drugs used in cardiology act by chang-ing the gating properties of ion channels. Thisis true for Cal+ antagonists which are blockersof voltage-operated L-type Ca2+ channels andfor most antiarrhythmics. Class I agents areblockers of Na+ channels whereas class IIIantiarrhythmics prolong the cardiac actionpotential by blockade of K+ channels.' In thepast decade much has been learnt about thephysiology and pharmacology of ion channels.Pharmacological agents that selectively openK+ channels have been developed and thisnovel group of drugs is rapidly expanding.23K+ channel openers are peripheral and coro-nary vasodilators45 but they additionally act onthe myocardium and seem to protect it fromdamage during ischaemia and reperfusion.67The purpose of this paper is to review some

of the pharmacology of cardiovascular K+channels and to elucidate the potential clinicalvalue of K+ channel opening as a new conceptin cardiovascular medicine.

Division of Cardiology,Medical DepartmentB, Rigshospitalet,University ofCopenhagen, DenmarkJ E Nielsen-KudskS BoesgaardJ AldershvileCorrespondence to:Dr J E Nielsen-Kudsk,Department of MedicineB2142, Rigshospitalet,University of Copenhagen,Blegdamsvej 9, DK-2100Copenhagen, Denmark.

Accepted for publication7 February 1996

Basic aspects of K+ channelsK+ CHANNELS AND MEMBRANE POTENTIALK+ channels play a key role in the regulation ofmembrane potential and cell excitability andthe function of these channels contributes tothe electrical and mechanical properties of theheart and vasculature. The concentration ofK+ inside cells (150 mM) is much higher thanoutside (3-5 mM) due to the action of theNa+/K+ pump. Opening of K+ channels makesK+ ions flow out of the cell along the outwarddirected electrochemical gradient for K+. Thischanges the membrane potential in a hyperpo-larising direction. It becomes more negativeand is moved towards the K+ equilibriumpotential. In contrast, blockade of K+ channelsshifts the membrane potential in a depolaris-ing direction. In the heart, repolarisation ofthe cardiac action potential is largely causedby opening of K+ channels.' Membrane depo-larisation and hyperpolarisation throughblockade and opening of K+ channels are alsoimportant mechanisms regulating vascularsmooth muscle contraction and relaxation.8

DIVERSITY OF K+ CHANNELSK+ channels are the most heterogeneous of allion channels. So far at least 16 major types ofK+ channels have been characterised and sev-

eral subtypes exist within each major type ofK+ channel. Each type of K+ channel serves adistinct function and the expression of K+channels differs among tissues and organs.This high degree of diversity opens the fasci-nating possibility that pharmacologically selec-tive openers or blockers of a specific K+channel subtype may be developed to modu-late specialised tissue functions. Identificationof the different types of K+ channels have beenmade possible largely because of the develop-ment of the electrophysiological patch-clampmethod which allows ion channels to be inves-tigated at a single channel level. By thismethod it is possible to determine factorswhich open and close a channel (the gatingproperties) as well as the single channel con-ductance which is a measure ofhow easily ionsflow through a channel (measured in picoSiemens). These variables are used to charac-terise and classify specific types of K+ chan-nels. K+ channels can be divided into threemajor classes on basis of their gating proper-ties: those that are gated by ligands such asATP, Ca2+, neurotransmitters or G-proteins(ligand gated), those gated by changes in mem-brane potential (voltage gated) and those gatedboth by ligands and voltage. The voltage-gatedK+ channels can be subdivided into outward ordelayed rectifiers which are activated by cellmembrane depolarisation and inward rectifierswhich are closed by depolarisation.

Several K+ channels have now been clonedthus allowing the molecular and structuralbasis of the diversity of K+ channels to bedetermined.9 On the basis of molecular struc-ture two major super-families of K+ channelscan be discriminated. The S4 superfamily (fig1), to which voltage-gated delayed rectifier K+channels belong, is built up of large proteinsubunits of six membrane-spanning a-helicalsegments (S 1-S6) and a seventh hairpin-shaped segment interposed between S5 andS6. The seventh segment forms a major part ofthe K+ channel pore itself (P-region) while S4 isbelived to be the voltage sensor. K+ channelsare tetramers formed by four subunits. Severaldifferent genes code for different K+ channelsubunits and these assemble as heteromulti-mers, resulting in an enormous diversity. Theinward rectifier K+ channel superfamily (fig 1)is characterised by subunits composed of onlytwo membrane spanning segments (Ml and

109

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from

Nielsen-Kudsk, Boesgaard, Aldershvile

Extracellular

Intracellular

Extracellular

Figure 1 The presumed structure ofK' channel subunits belonging to the S4 superfamily(upper) and the inward rectifier superfamily (lower) ofK' channels.

M2). The pore region is shaped like a hairpinand is almost identical to that of the S4 super-

family.

CARDIOVASCULAR K+ CHANNELS

The most important cardiovascular K+ chan-nels are listed in table 1. The outward rectifieror delayed rectifier (Kv) channel is activatedby membrane depolarisation and gives rise toan outward directed K+ current which isresponsible for repolarisation of the cardiacaction potential. It can be blocked by class IIIantiarrhythmic agents. The inward rectifier

(K,,I) keeps the resting membrane potentialstable during cardiac diastole. It is open at theresting membrane potential but closes duringdepolarisation. This type of channel is sparse

in conduction tissue that undergoes sponta-neous depolarisation in diastole. The acetyl-choline activated K+ channel (KAch) isactivated by the binding of acetylcholinereleased from vagal nerve endings to mus-

carinic receptors in atrial and atrioventricular(AV) nodal tissue. This channel mediatesbradycardia and the decrease in AV-nodalconduction produced by vagal stimulation.The large conductance Ca2+-activated K+channel (BKla) acts as a negative feedbackmechanism in vascular smooth muscle. It isactivated when intracellular Ca2+ is increasedby the action of vasoconstrictors and it limitsdepolarisation and excessive Ca2+ stimulation.The ATP-sensitive K+ channel (KATP) is inhib-ited by high intracellular ATP concentrations.It is the target K+ channel subtype for syn-

thetic K+ channel openers and is consideredbelow.

The ATP-sensitive K+ channelThe KATP channel was first identified in car-

diac myocytes.'0 It is believed to have a struc-ture similar to the inward rectifyingsuperfamily of K+ channels. This type of chan-nel is inhibited by a high intracellular concen-

tration of ATP and is closed under conditionsof normal myocardial metabolism. The chan-nel opens under conditions of myocardialischaemia. Channel opening is stimulated by a

fall in ATP concentration and by acidosis, lac-tate, adenosine, and nucleotide diphosphatessuch as ADP and GDP. These factors may actin concert to activate the channel duringmyocardial ischaemia."' The consequence ofKATP opening is increased K+ efflux, acceler-ated repolarisation, and thereby a shorteningof the cardiac action potential. Shortening ofthe plateau phase reduces the time availablefor Ca2+ influx which causes a decline in con-

tractile function in the ischaemic zone. Theseeffects reduce the energy consumption, sparehigh energy phosphates, and limit Ca2+ over-

load and hence tend to prolong survival of theischaemic tissue. In this way KATP seems torepresent an endogenous protective mecha-nism against myocardial metabolic stress.'0 12

KATP channels have also been identified in

Table 1 Major types of cardiovascular K+ channels

ConductanceType Class (pS) Factors which induce opening Blockers Structure Function

KATP Ligand gated 10-30 Decreased [ATP]i, increased [NDP],, High [ATP]i, IR Hyperpolarisation during(ATP sensitive) K+ channel openers-eg cromakalim, glibenclamide, metabolic stress and

pinacidil, nicorandil, minoxidil sulphate 5-hydroxydecanoate hypoxia/ischaemiaKAChI ILigand gated 10-25 G-proteins through muscarinic receptor IR Bradycardia by vagus(ACh activated) stimulation stimulationKV Voltage gated 5-15 Depolarisation, specific openers not yet 4-Aminopyridine S4 Repolarisation of cardiac(delayed rectifier) available action potentialKIR Voltage gated 10-15 Hyperpolarisation Depolarisation IR Maintenance of resting(inward rectifier) potentialBK-a Ligand and 100-150 Increased [Ca2+] , depolarisation, Charybodotoxin, 84 Reduction of excitability(Ca2+ activated) voltage gated specific openers not yet available iberotoxin, when [Ca2+]i rises

TEA (1 mM)

NDP, nucleotide diphosphates such as adenosine diphosphate (ADP), guanidine diphosphate (GDP), and uridine diphosphate (UDP); ACh, acetylcholine; IR andS4, inward rectifier and S4 superfamilies of K+ channels.

110

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from

K+ channel opening: a new drug principle in cardiovascular medicine

Coronary smooth muscle cell

HyperpolarisationVasorelaxation

Coronary blood flow+

Figure 2 KATP channels open during ischaemia. They are stimulated by acidosis, a fall inintracellular A TP, a rise in nucleotide diphosphates (NDP) such as ADP and GDP, andby accumulation of extracellular adenosine. Adenosine activates KATP by a G-proteindependent mechanism after stimulation of adenosine receptors (A) on the cell surface.Potassium channel openers (PCOs) are pharmacological activators ofKATP whereas theantidiabetic sulfonylurea glibenclamide (GLI) is a specific blocker. Some of theconsequences ofKATP activation in the ischaemic myocardium and coronary vasculatureare listed. APD, action potential duration. (Adapted with permission from Escande D,Cavero I. Potassium channel openers in the heart. In: Escande D, Standen N, eds.K+ Channels in cardiovascular medicine. Paris: Springer-Verlag, 1993;225-446.)

vascular smooth muscle.'3 Like the cardiacKATP this channel is inhibited by ATP andstimulated by nucleotide diphosphates. Itopens under conditions of hypoxia and KATPopening may be one of the principal mecha-nisms which underlie hypoxic coronaryvasodilation.14 Figure 2 shows some of theimportant factors which gate the KATP inmyocardial and coronary vascular smoothmuscle cells.6

Table 2 Selected KATPopeners

Benzopyrans:CromakalimLevcromakalim (active

enantiomer ofcromakalim)

BimakalimHOE234

Cyanoguanidines:PinacidilP1705BMS-1 80448

Thiformamides:AprikalimRP49356 (active

enantiomer ofaprikalim)

Pyrimidines:Minoxidil sulphate

Benzothiadiazines:Diazoxide

Nitrate containing:NicorandilKRN2391

ENDOGENOUS MODULATION OF KArPAdenosine opens KATP both in cardiomy-ocytes'5 and coronary vascular smooth musclecells'6 probably through a G-protein depen-dent mechanism triggered by adenosinereceptor stimulation. Other endogenousvasodilators such as calcitonin gene regulatedpeptide (CGRP), vasoactive intestinal peptide(VIP), endothelium derived hyperpolarisingfactor (EDHF), and prostacyclin act in partthrough KATP opening.'3 Most recently it hasbeen reported that nitric oxide (endotheliumderived relaxing factor, EDRF) apart from itsstimulating effect on guanylate cyclase directlyactivates K+ channels of the BKcl type.'7 Incultured coronary artery smooth muscle cellsthe vasoconstrictors angiotensin II andendothelin inhibit KATPI,3 but the functionalsignificance of these findings is not yet clear.

PHARMACOLOGICAL MODULATION OF KAl,,The best known of the new KATP channelopening drugs are cromakalim, levcromakalim(the active enantiomer of cromakalim),pinacidil, nicorandil, and aprikalim. Thechemical structures of these compounds are

quite different and they have been subdividedinto different groups (table 2). The first agentrecognised as a K+ channel opener was nico-

randil,'8 which is now in clinical use as anantianginal drug. However, this drug also con-tains a nitrate moiety and has a nitrovasodilatoraction in addition to its ability to open K+channels. Cromakalim, which was developedas an arterial vasodilator, was the first specificsynthetic K+ channel opener.2 4 The mecha-nism of action of this drug was reported in1986. Later, some of the classic vasodilatordrugs such as minoxidil sulphate (the activemetabolite of minoxidil) and diazoxide werefound to act by K+ channel opening.24 At pre-sent, all available synthetic K+ channel openersseem to be openers of KATP. There have beenattemps to develop openers of the BKCa channel(SCA40 and NS16 19) but these agents are notselective. However, it is likely that selectivepharmacological modulators for specific K+channel subtypes will be developed in thefuture.The oral antidiabetic sulfonylurea drugs are

selective blockers of KATP,2 4 and glibenclamideis now used as an important pharmacologicaltool to detect actions produced by drugs orendogenous substances which involve KATPopening. KATP channels are also present in pan-creatic islets where they are essential to thephysiological regulation of insulin secretion.'9In this tissue the KATP channels are open underconditions of normal blood glucose but closewhen the intracellular concentration of ATP israised by increased concentrations of glucose.Closure of KATP leads to islet cell depolarisationand insulin secretion. Glibenclamide and othersulfonylureas specifically block KATP and in thisway promote insulin secretion. Concentrationsof glibenclamide in the nM range are sufficientto block pancreatic KATP channels whereas ,Mconcentrations are required to block cardiovas-cular KATP channels. With the exception of dia-zoxide, very high concentrations of the K+channel openers are required to open pancre-atic KATP and inhibit insulin secretion.Diazoxide does not discriminate between car-diovascular and pancreatic KATP channels andbecause of its hyperglycaemic action it is nolonger in clinical use. Minoxidil sulphate whichis another opener of cardiovascular KATP chan-nels, actually blocks pancreatic KATP. Thus KATPis not a single distinct type of channel but rathera family of several KATP subtypes with importanttissue-dependent differences in sensitivity topharmacological modulators.

Effects of K+ channel openers on thecirculationSYSTEMIC CIRCULATIONK+ channel openers have been shown to relaxisolated systemic arteries contracted by a widerange of vasoconstrictors.5 These drugs are10-100 times more potent on vascular smoothmuscle than on the non-ischaemic myo-cardium.6 A unique characteristic of thesedrugs is their ability to relax contractionsinduced by moderately raised extracellular K+concentrations (20-30 mM) but not thoseinduced by high concentrations of K+ in the40-120 mM range.4 As the extracellular K+concentration is raised the electrochemical

Cardiac myocyte

APD shorteningHypokinesia/akinesiaCardioprotection

ill

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from

Nielsen-Kudsk, Boesgaard, Aldershvile

gradient for K+ is reduced and as a conse-

quence the K+ efflux produced by K+ channelopening becomes gradually diminished. Thecellular events20 which lead to vasorelaxationare shown in fig 3. The vasodilatation pro-duced both in vitro and in vivo by these drugscan be blocked by glibenclamide.45The haemodynamic profile of K+ channel

openers in various animal species and man ischaracteristic of potent arterial vasodilators.5They cause reflex stimulation of the sympa-

thetic nervous system and of the renin-angiotensin system. Even at doses whichgreatly reduce blood pressure these drugs donot have a significant cardiodepressant action.

PULMONARY CIRCULATIONVasoconstriction in response to hypoxia is a

unique property of pulmonary arteries. It con-

strasts the hypoxic vasodilatation seen in coro-

nary and systemic arteries. Pulmonary arteriesdepolarise during hypoxia and may even dis-charge action potentials.2' Evidence is accu-

mulating that K+ channel blockade is a keyevent which links hypoxia to pulmonary vaso-

constriction.2223 Exactly which type of K+channel is involved is not clear. K+ channelopeners relax isolated pulmonary arteries andsome of these drugs have more potent relaxanteffects in pulmonary arteries than in systemicarteries.24 In addition they inhibit hypoxic pul-monary vasoconstriction in animals.2' In a

study of acute haemodynamics in patients withangina pectoris, cromakalim reduced pul-monary vascular resistance to the same extentas systemic vascular resistance.25 Thus K+

Figure 3 Cellularmechanisms involved invascular smooth musclerelaxation induced bypotassium channelopening. VOC, voltage-operated Ca2+ channel.

Potassium channelopening

Potassium efflux

Vascular smoothmuscle relaxation

channel openers have a direct pulmonaryvasodilator action in addition to their effect onthe systemic circulation.

Effects ofK+ channel openers on the heartCORONARY CIRCULATION

KATP channels seem to be important in regulat-ing the coronary circulation. In isolated per-fused guinea pig hearts the coronaryvasodilatation induced by hypoxia andischaemia was prevented by glibenclamide andmimicked by cromakalim, indicating thatopening of KATP may be important in hypoxicand ischaemic coronary vasodilatation.'4 Theactivation of KATP may result from a directeffect of hypoxia or ischaemia on coronary vas-cular smooth muscle cells,26 but also fromadenosine released from the surroundingmyocardium. In the dog coronary circulationglibenclamide reduced both the reactivehyperaemic response after a brief coronaryocclusion27 and the autoregulatory vasodila-tion in response to low perfusion pressure.28 Inaddition glibenclamide causes vasoconstric-tion during basal coronary flow29 suggestingthat KATP may have several important regula-tory actions in the coronary circulation.

Pharmacological openers of KATP are potentcoronary vasodilators in vitro and in vivo.Regional haemodynamic studies have shownthat these drugs preferentially produce coro-nary vasodilatation.5 The K+ channel openeraprikalim even produces coronary vasodilata-tion in doses that have no effect on blood pres-sure.5 In vitro these drugs produce coronaryvasorelaxation when the endothelium is presentor when it is removed. Cromakalim, pinacidil,and nicorandil dilate large epicardial coronaryarteries as well as small coronary resistancearteries in vivo. However, damage of theendothelium by balloon angioplasty in dogsgreatly reduced the dilatation of large coronaryarteries by cromakalim and pinacidil whereasthat produced by nicorandil was unaffected.30Cromakalim and pinacidil therefore seem todilate large coronary arteries in vivo partly byan endothelium-dependent mechanism possi-bly induced by an increase in proximal flowcaused by dilatation of the small coronary resis-tance vessels. The endothelium-independentcoronary vasodilatation of large coronary arteriesproduced by nicorandil is probably mediated bythe additional nitrate action of the drug.

MYOCARDIUMK+ channel openers only influence the func-tion of the non-ischaemic myocardium at con-centrations that are much higher than thoserequired for vascular smooth muscle relax-ation. At such high concentrations these drugsshorten cardiac action potential duration anddepress cardiac contractility.6 However, atmuch lower concentrations these drugs exertcardioprotective actions, with some KATPopeners acting at concentrations which pro-duce only minimal haemodynamic effects.3' Inglobally ischaemic rat hearts, cromakalimand pinacidil improved the post-ischaemicrecovery of contractile function and cro-

12

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from

K+ channel opening: a new drug principle in cardiovascular medicine

makalim additionally reduced lactate dehydro-genase release.32 At cardioprotective concen-trations, the K+ channel openers did notinfluence the contractile function of non-ischaemic hearts. Several other KATP openershave proved cardioprotective in isolated heartmodels. In animal models of myocardial stun-ning, K+ channel openers markedly improvedthe recovery of myocardial function afterischaemia." In general these drugs are onlycardioprotective when administered beforereperfusion; they are ineffective if dosing isdelayed and initiated during reperfusion."Also in animal models of myocardial infarctionKATP openers seem to be effective in reducingmyocardial infarct size.32The precise mechanism underlying the car-

dioprotective effects of K+ channel openers isnot yet clarified. Several studies have demon-strated that glibenclamide blocks the anti-ischaemic effects which therefore seem to bemediated by opening of KATP.'2 Another conse-quent finding is that cardioprotection can beachieved without coronary vasodilatation ordilatation of coronary collaterals, indicatingthat the effect is not dependent upon coronaryvascular smooth muscle relaxation butinvolves a direct myocardial effect.32 This isfurther supported by the development of KATPopeners such as BMS-180448 that are selec-tive for the ischaemic myocardium with ahigher cardioprotective than vasorelaxantpotency.34 The fact that these drugs have to beadministered before reperfusion to be effec-tive, indicates that they do not inhibit reperfu-sion injury directly but rather are protectiveduring ischaemia itself. A mechanism that mayaccount for the cardioprotective effect of KATPopeners is the accelerated shortening of car-diac action potential duration in earlyischaemia caused by an enhanced K+ outwardcurrent. This would limit Ca2+ influx, counter-act intracellular Ca2+ overload, enhance thedecline of contractile function, and shorten thetime to arrest of mechanical activity in theischaemic zone of the myocardium."" 12 35 Such acardioplegic mechanism would save energyand tend to prolong survival of the cardiomy-ocytes. Experiments have shown that cardio-protection by KA,T openers is associated withpreservation of myocardial ATP.'2 However,in recent studies in dogs with bimakalim andcromakalim there seemed to be some dissocia-tion between action potential shortening andcardioprotection, indicating that additionalmechanisms may be involved.'637 Duringreperfusion after global ischaemia in rathearts, cromakalim improved oxygen effi-ciency; this suggests a direct effect on cardiacmetabolism, possibly at the level of the mito-chondria.32

Ischaemic preconditioning is a fascinatingendogenous cardioprotective mechanism ofthe heart. In this phenomenon one or morebrief periods of coronary occlusion markedlyreduce the size of the myocardial infarctionthat follows a subsequent prolonged coronaryartery occlusion. There is now substantialexperimental evidence suggesting thatischaemic preconditioning is mediated partly

or in whole by opening of KATP.'8 Gross andAuchampach39 first showed that glibenclamideabolished the cardioprotective effect of pre-conditioning in dogs and that a KATP openeraprikalim mimicked the effect of precondition-ing without producing any systemic haemody-namic effects. Later studies in dogs and pigssupported an essential role for KATP in precon-ditioning, whereas results of studies in rabbitsand rats were somewhat conflicting.'8 In arecent study by Yao and Gross,40 transientocclusion of the left anterior descending coro-nary artery (LAD) for only 3 min had nopreconditioning effect. However, whenbimakalim was given during the LAD occlu-sion period a significant preconditioning effectwas seen. Bimakalim infusion alone did notinfluence infarct size, thus suggesting thatpharmacological activation of KATP can sensi-tise the myocardium to preconditioning.Glibenclamide also blocks preconditioningin man.4' In patients with angina pectoris,coronary angioplasty with two subsequentballoon inflations resulted in markedlyreduced ST segment changes and pain duringthe second balloon inflation. The beneficialpreconditioning effect of the first inflationwas completely abolished by glibenclamidegiven in an oral dose used in antidiabetic treat-ment.

Previous studies have also suggested a rolefor adenosine in mediating ischaemic precon-ditioning. Adenosine acts through adenosinereceptor stimulation in the myocardium andactivates KATP by a G-protein regulated mecha-nism.'5 Thus it seems likely that KATP opening isthe mechanism whereby adenosine producesits protective effect. This is supported by theobservation that glibenclamide in dogs com-pletely blocked the ability of exogenouslyadministered adenosine to mimic precondi-tioning.42

ELECTROPHYSIOLOGICAL ASPECTSConcern has been raised that K+ channelopeners may possibly be proarrhythmicbecause of their ability to shorten cardiacaction potential duration (APD), an effect thatis expected to reduce cardiac refractoriness.However, shortening of cardiac APD may beboth proarrhythmic and antiarrhythmicdepending on the type of arrhythmia in ques-tion and the state of the myocardium. As aconsequence there are experimental studiesreporting proarrhythmic, not proarrhythmic,and antiarrhythmic effects with thesedrugs.4345 In the non-ischaemic myocardium,only very high and clinically irrelevant concen-trations produce shortening ofAPD and a ten-dency to arrhythmias. In contrast, clinicallyrelevant concentrations inhibit arrhythmiasinduced by triggered activity (early and lateafterdepolarisations) which are of relevance inthe long QT syndrome and torsades depointes ventricular tachycardia. In theischaemic myocardium, which favours reentryarrhythmias, these drugs shorten cardiac APDat clinically relevant concentrations and mayfacilitate arrhythmias. Whereas several studieshave shown no proarrhythmia or even antiar-

113

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from

Nielsen-Kudsk, Boesgaard, Aldershvile

rhythmic activity of K+ channel openers in ani-mal models of ischaemia-induced arrhythmiasothers have shown that proarrhythmias may bea risk particularly at the high concentrationsthat produce profound hypotension. In thiscontext, K+ channel openers by their cardio-protective action may produce indirect antiar-rhythmic effects in myocardial ischaemia. Todate, pinacidil and nicorandil have been givento thousands of patients without reports ofadverse arrhythmias. If the cardioprotectiveeffect can be separated from the ability ofthese drugs to shorten cardiac APD, and someobservations have suggested that they can,37 itshould be possible to develop cardioprotectiveKATP openers without or with only a minimalproarrhythmic potential.The ST segment and T wave changes in

the electrocardiogram during myocardialischaemia are believed to result from enhancedoutwardly directed K+ currents. K+ channelopeners may therefore mimic such changeseven in the absence of myocardial ischaemia.Changes consisting of inversion or flatteningof the T wave have been reported in 30% ofpatients treated with pinacidil.46 Even thoughsuch changes are benign, they may give rise todifficulties in the interpretation of the electro-cardiogram, particularly in patients withischaemic heart disease.

Therapeutic potential of K+ channelopenersANGINA PECTORISCoronary vasodilatation induced by hypoxia/ischaemia is an endogenous beneficial mecha-nism which operates to improve perfusion ofthe ischaemic myocardium. The role of KA,Pin mediating metabolic coronary vasodilata-tion makes KATP openers of potential interestin the treatment of angina pectoris. In addi-tion, these vasorelaxant drugs preferentiallyproduce coronary vasodilatation, some ofthem even without systemic haemodynamicchanges. Their ability to inhibit myocardialstunning, limit myocardial infarct size, andmimic ischaemic preconditioning may furtherstimulate the interest in these drugs in coro-nary artery disease. However, although clinicaltrials have been initiated, there are currentlyno clinical data on pure K+ channel openers inpatients with ischaemic heart disease.

Clinical data are available for nicorandil47 48which is now in clinical use as an antianginalagent. Although it is classified as a K+ channelopener it is not a pure one, but a hybridbetween a nitrate and a KATP opener.Nicorandil has proved effective for the treat-ment of stable angina pectoris in placebo con-trolled clinical trials, and in comparativestudies it shows the same degree of antianginalefficacy as nitrates, Ca2+ antagonists, and ,Badrenoceptor blockers. In addition, clinicaldata indicate that it might be useful in thetreatment of unstable angina and variantangina.

Nicorandil dilates large coronary arteriesand has balanced preload and afterload reduc-ing effects because of its nitrate and KATP

opener components, respectively. It increasescardiac output by haemodynamic changes andmay be used in patients with poor left ventricu-lar function. Interestingly nicorandil, unlikethe nitrates, does not seem to induce thedevelopment of tolerance.

CARDIOPROTECTIONThe cardioprotective effect of KATP openers isa most fascinating aspect of these new drugs.It implies that they may be able not only toprevent and relieve the symptoms of myocar-dial ischaemia but also to reduce some ofthe important consequences of myocardialischaemia such as myocardial stunning andinfarction. If this holds true in clinical prac-tice, it is likely that these drugs will be of long-term benefit and reduce mortality in patientswith ischaemic heart disease. These drugs mayalso have a therapeutic potential as adjuvanttreatment in coronary angioplasty and coro-nary bypass surgery to protect themyocardium from ischaemic injury and toinhibit postoperative myocardial stunning.

HEART FAILUREHybrid drugs between nitrates and KATP open-ers such as nicorandil have a balanced haemo-dynamic profile of preload and afterloadreduction.48 This effect profile resembles thatproduced by the combined treatment ofisosorbide dinitrate and hydralazine which wasused in the V-HeFT studies. In the V-HeFT IIstudy the ACE inhibitor enalapril was superiorto isosorbide dinitrate and hydralazine interms of mortality, whereas the latter combi-nation was superior to enalapril in improvingleft ventricular function and exercise capacityof the patients. Nicorandil has so far shownfavourable haemodynamic actions in heartfailure in small short-term clinical studies.48

HYPERTENSIONK+ channel openers were originally developedas antihypertensive vasodilators. The beststudied drug is pinacidil,46 which is in clinicaluse for the treatment of hypertension. Thisdrug is clearly effective in controlling hyper-tension, but as with other directly actingperipheral vasodilatators it causes reflex tachy-cardia and fluid retention and is not suited tomonotherapy. It can be used in combinationwith /B adrenoceptor blockers, diuretics, orACE inhibitors. It has no adverse effect onblood lipids, glucose tolerance, or insulinsecretion. Other K+ channel openers such aslevcromakalim have been selected for develop-ment as antihypertensive agents.

PULMONARY HYPERTENSIONA clinically useful vasodilator for the treatmentof pulmonary hypertension should be selectivefor the pulmonary circulation and producepulmonary vasodilatation without significantsystemic hypotension and cardiodepression.Hypoxic vasoconstriction is a unique featureand probably the most powerful control mech-anism of vascular tone in the pulmonary circu-lation.2' The essential role of inhibition of K+channels in this regulatory mechanism raises

14

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from

K+ channel opening: a new drug principle in cardiovascular medicine

the possibility that pharmacological openers ofthe K+ channel involved might be selectivepulmonary vasodilators. Although the K+channel subtype involved appears to be differ-ent from KATP, openers of this channel pro-duce pulmonary vasodilatation in vitro and invivo and inhibit hypoxic pulmonary vasocon-striction in animals.23 Pinacidil and relatedcyanoguanidine derivatives were more potentrelaxants of pulmonary than systemic arteriesfrom the guinea pig.24 So far there are no clini-cal data available on the effect of K+ channelopeners in patients with pulmonary hyper-tension.

PERIPHERAL VASCULAR DISEASEVasodilators have been unsuccessful in thetreatment of intermittent claudication proba-bly because they dilate vessels supplying nor-mally perfused skeletal muscle and therebydivert blood flow away from the ischaemic tis-sue. In contrast to other vasodilators K+ chan-nel openers seem to enhance blood flow toischaemic skeletal muscle and improve recov-ery of muscle energy stores in animal modelsof chronic occlusive arterial disease.49 Thesebeneficial effects were manifest at doses belowthose affecting systemic blood pressure andmay reflect redistribution of blood flow toischaemic muscle due to selective dilatation ofcollateral vessels. Hypersensitivity of such ves-sels to K+ channel openers might arise becauseischaemic conditions would favour opening ofKATP. Another possible explanation could bedifferences in nerve supply and populations ofK+ channels between collaterals and other ves-sels. Clinical studies on the effect of K+ channelopeners in patients with peripheral vasculardisease have started.

IMPLICATIONS FOR DIABETIC PATIENTSTREATED WITH SULPHONYLUREA KA,PBLOCKERSDiabetes mellitus is an adverse risk factor inacute myocardial infarction. These patientshave a poor prognosis after infarction with athree to four fold higher mortality risk thannon-diabetics. The reason why diabeticpatients do so badly is unclear. Non-insulin-dependent diabetes mellitus (NIDDM)accounts for 85% of all cases of diabetes andmost of these patients are treated with oralantidiabetic sulfonylureas which can block notonly pancreatic KATP channels but also cardio-vascular KATP channels. The question ariseswhether treatment with antidiabetic sulfony-lureas is hazardous from a cardiovascularpoint of view. In a large multicentre trial withan eight year observation period a twofoldincrease in cardiovascular mortality wasobserved among patients treated with tolbu-tamide when compared with placebo orinsulin treated patients,90 but the validity ofthis conclusion has been questioned and is thesubject of much debate.9' If the knowledgederived from basic science of the interactionbetween antidiabetic sulfonylureas and cardio-vascular KATP channels is clinically relevant theuse of sulfonylureas in patients with diabetesshould be considered carefully.

ConclusionK+ channel openers are a group of novel drugsthat target KATP channels in the cardiovascularsystem. This type of K+ channel opens duringmyocardial ischaemia and plays a part inimportant endogenous protective mechanismssuch as hypoxic/ischaemic coronary vasodi-latation and ischaemic preconditioning. Byexploiting these natural protective mecha-nisms of the heart these drugs inhibit myocar-dial stunning, limit myocardial infarct size,and sensitise the myocardium to ischaemicpreconditioning in animal experiments. Thecardioprotective effects of the drugs are pro-duced at low concentrations which have noeffects in the non-ischaemic myocardium.Most of the drugs are also peripheral vasodila-tors and some of the first drugs developed arein clinical use as antihypertensive agents. KATPopeners which are selective for the ischaemicmyocardium have been synthesised and seemto have an interesting therapeutic potential inischaemic heart disease. Clinical trials havestarted, but so far there is no clinical experi-ence with pure KATP openers in ischaemicheart disease. Nicorandil, which is a hybridbetween a nitrate and a KATP opener, is in clin-ical use as an effective agent for the treatmentof angina pectoris. Pulmonary hypertensionand peripheral vascular disease are otherpotential clinical applications for this novelpharmacological concept of opening cardio-vascular K+ channels.

The authors are receiving grants from the Danish HeartFoundation, the J0rgen Moller Foundation, the ResearchFoundation of NOVO, and the Foundation of Krista andViggo Petersen.

1 Katz AM. Cardiac ion channels. N Engl J Med 1993;328:1244-51.

2 Quast U. Potassium channel openers: pharmacological andclinical aspects. Fundam Clin Pharmacol 1992;6:279-93.

3 Edwards G, Weston AH. Pharmacology of the potassiumchannel openers. Cardiovasc Drugs Ther 1995;9:185-93.

4 Edwards G, Duty S, Trezise DJ, Weston AH. Effects ofpotassium channel modulators on the cardiovascular sys-tem. In:. Weston AH, Hamilton TC, eds. Potassium chan-nel modulators. Oxford: Blackwell Scientific Publications,1992;369-421.

5 Cavero I, Guillon JM. Pharmacological profile of potas-sium channel openers in the vasculature. In: Escande D,Standen N, eds. K+ Channels in cardiovascular medicine.Paris: Springer-Verlag, 1993; 193-223.

6 Escande D, Cavero I. Potassium channel openers in theheart. In: Escande D, Standen N, eds. K+ Channels in car-diovascular medicine. Paris: Springer-Verlag, 1993;225-44.

7 Cavero I, Djellas Y, Guillon JM. Ischemic myocardial cellprotection conferred by the opening of ATP-sensitivepotassium channels. Cardiovasc Drugs Ther 1995;9:245-55.

8 Nelson MT, Quayle JM. Physiological roles and propertiesof potassium channels in arterial smooth muscle. Am _7Physiol 1995;268:C799-C822.

9 Kukuljan M, Labarca P, Latorre R. Molecular determi-nants of ion conduction and inactivation in K+ channels.Am j'Physiol 1995;268:C535-C56.

10 Noma A. ATP-regulated K+ channels in cardiac muscle.Nature 1983;305:147-8.

11 Coetzee WA. ATP-sensitive potassium channels andmyocardial ischaemia: why do they open? CardiovascDnrgs Ther 1992;6:201-8.

12 Escande D, Cavero I. K+ channel openers and 'natural' car-dioprotection. Trends Pharmacol Sci 1992;13:269-72.

13 Quayle JM, Standen NB. KATP channels in vascular smoothmuscle. Cardiovasc Res 1994;28:797-804.

14 Daut J, Maier Rudolph W, von Beckerath N, Mehrke G,Gunther K, Goedel-Meiner L. Hypoxic dilation of coro-nary arteries is mediated by ATP-sensitive potassiumchannels. Science 1990;247:1341-4.

15 Kirsch GE, Codina J, Bimbaumer L, Brown AM. Couplingof ATP-sensitive K+ channels to A, receptors by G pro-teins in rat ventricular myocytes. Am J Physiol 1990;259:H820-H6.

115

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from

Nielsen-Kudsk, Boesgaard, Aldershvile

16 Merkel LA, Lappe RW, Rivera LM, Cox BF, Perrone MH.Demonstration of vasorelaxant activity with an A,-selec-tive adenosine agonist in porcine coronary artery:involvement of potassium channels. J7 Pharmacol Exp Ther1992;260:437-43.

17 Bolotina VM, Najibi S, Palacino JJ, Pagano PJ, Cohen RA.Nitric oxide directly activates calcium-dependent potas-sium channels in vascular smooth muscle. Nature 1994;368:850-3.

18 Furukawa K, Itoh T, Kajiwara M, Kitamura K, Suzuki H,Ito Y, et al. Vasodilating actions of 2-nicotinamidoethylnitrate on porcine and guinea-pig coronary arteries.Pharmacol Exp Ther 1981;218:248-59.

19 Schmid Antomarchi H, De Weille J, Fosset M, LazdunskiM. The receptor for antidiabetic sulfonylureas controlsthe activity of the ATP-modulated K+ channel in insulin-secreting cells. Biol Chem 1987;262:15840-4.

20 Quast U, Guillon JM, Cavero I. Cellular pharmacology ofpotassium channel openers in vascular smooth muscle.Cardiovasc Res 1994;28:805-10.

21 Barnes PJ, liu SJ. Regulation of pulmonary vascular tone.Pharmacol Rev 1995;47:87-131.

22 Post JM, Hume JR, Archer SL, Weir EK. Direct role forpotassium channel inhibition in hypoxic pulmonary vaso-constriction. Am Physiol 1992;262:C882-C90.

23 Kozlowski RZ, Nye PCG. Potassium channels in the pul-monary vasculature. In: Escande D, Standen N, eds. K'channels in cardiovascular medicine. Paris: Springer-Verlag,1993; 141-58.

24 Nielsen-Kudsk JE, Bang L. Effects of pinacidil and othercyanoguanidine derivatives on guinea-pig isolated tra-chea, aorta and pulmonary artery. EurJ Pharmacol 1991;201:97-102.

25 Thomas P, Dixon MS, Winterton SJ, Sheridan DJ. Acutehaemodynamic effects of cromakalim in patients withangina pectoris. BrJ Clin Pharmacol 1990;29:325-31.

26 Mellemkjaer S, Nielsen-Kudsk JE. Glibenclamide inhibitshypoxic relaxation of isolated porcine coronary arteriesunder conditions of impaired glycolysis. Eur Pharmacol1994;270:307-12.

27 Aversano T, Ouyang P, Silverman H. Blockade of theATP-sensitive potassium channel modulates reactivehyperemia in the canine coronary circulation. Circ Res1991;69:618-22.

28 Komaru T, Lamping KG, Eastham CL, Dellsperger KC.Role of ATP-sensitive potassium channels in coronarymicrovascular autoregulatory responses. Circ Res 1991;69:1146-51.

29 Imamura Y, Tomoike H, Narishige T, Takahashi T,Kasuya H, Takeshita A. Glibenclamide decreases basalcoronary blood flow in anesthetized dogs. Am Physiol1992;263:H399-H404.

30 Ghaleh B, Dubois Rande JL, Hittinger L, Giudicelli JF,Berdeaux A. Comparisons of the effects of nicorandil,pinacidil, nicardipine and nitroglycerin on coronary ves-sels in the conscious dog: role of the endothelium. Br YPharmacol 1995;114:496-502.

31 Auchampach JA, Maruyama M, Cavero I, Gross GJ. Thenew K+ channel opener aprikalim (RP 52891) reducesexperimental infarct size in dogs in the absence of hemo-dynamic changes. Y Pharmacol Exp Ther 1991;259:961-7.

32 Grover GJ. Protective effects of ATP-sensitive potassium-channel openers in experimental myocardial ischemia. YCardiovasc Pharmacol 1994;24 suppl 4:S18-S27.

33 Auchampach JA, Maruyama M, Cavero I, Gross GJ.

Pharmacological evidence for a role of ATP-dependentpotassium channels in myocardial stunning. Circulation1992;86:311-9.

34 Grover GJ, McCullough JR, D'Alonzo AJ, Sargent CA,Atwal KS. Cardioprotective profile of the cardiac-selective ATP-sensitive potassium channel opener BMS-180448..7 Cardiovasc Pharmacol 1995;25:40-50.

35 Mellemkjaer S, Nielsen-Kudsk JE. Effects of levcro-makalim and glibenclamide on paced guinea-pig atrialstrips exposed to hypoxia. Eur Jf Pharmacol 1995;277:51-6.

36 Yao Z, Gross GJ. Effects of the KATP channel openerbimakalim on coronary blood flow, monophasic actionpotential duration, and infarct size in dogs. Circulation1994;89: 1769-75.

37 Grover GJ, D'Alonzo AJ, Parham CS, Darbenzio RB.Cardioprotection with the KATP opener cromakalim is notcorrelated with ischemic myocardial action potentialduration. J Cardiovasc Pharcacol 1995;26: 145-52.

38 Gross GJ. ATP-sensitive potassium channels and myocar-dial preconditioning. Basic Res Cardiol 1995;90:85-8.

39 Gross GJ, Auchampach JA. Blockade of ATP-sensitivepotassium channels prevents myocardial preconditioningin dogs. Circ Res 1992;70:223-33.

40 Yao Z, Gross GJ. Activation of ATP-sensitive potassiumchannels lowers threshold for ischemic preconditioningin dogs. AmJ Physiol 1994;267:H1888-H94.

41 Tomai F, Crea F, Gaspardone A, Versaci F, Paulis RD, dePeppo AP, et al. Ischemic preconditioning during coro-nary angioplasty is prevented by glibenclamide, a selec-tive ATP-sensitive K+ channel blocker. Circulation 1994;90:700-5.

42 Grover GJ, Sleph PG, Dzwonczyk S. Role of myocardialATP-sensitive potassium channels in mediating precon-ditioning in the dog heart and their possible interactionwith adenosine A,-receptors. Circulation 1992;86: 1310-6.

43 Wilde AA, Janse MJ. Electrophysiological effects of ATPsensitive potassium channel modulation: implications forarrhythmogenesis. Cardiovasc Res 1994;28: 16-24.

44 Black SC, Lucchesi BR. Potassium channel openers arelikely to be proarrhythmic in the diseased human heart.Cardiovasc Res 1994;28:923-4.

45 D'Alonzo AJ, Grover GJ. Potassium channel openers areunlikely to be proarrhythmic in the diseased humanheart. Cardiovasc Res 1994;28:924-5.

46 Friedel HA, Brogden RN. Pinacidil. A review of its phar-macodynamic and pharmacokinetic properties, and thera-peutic potential in the treatment of hypertension. Drugs1990;39:929-67.

47 Frampton J, Buckley MM, Fitton A. Nicorandil. A reviewof its pharmacology and therapeutic efficacy in anginapectoris. Drugs 1992;44:625-55.

48 Knight C, Purcell H, Fox K. Potassium channel openers:clinical applications in ischemic heart disease-overviewof clinical efficacy of nicorandil. Cardiovasc Drugs Ther1995;9:229-36.

49 Cook NS, Chapman ID. Therapeutic potential of potas-sium channel openers in peripheral vascular disease andasthma. Cardiovasc Drugs Ther 1993;7 suppl 3:555-63.

50 University Group Diabetes Program. A story of the effects ofhypoglycemic agents on vascular complications inpatients with adult-onset diabetes. II. Mortality results.Diabetes 1970;19:789-830.

51 Smits P, Thien T. Cardiovascular effects of sulphonylureaderivatives. Implications for the treatment of NIDDM?Diabetologia 1995;38: 116-21.

116

on October 9, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.76.2.109 on 1 A

ugust 1996. Dow

nloaded from