Cardiovascular Research 1992;26:1011-1016
1011
Review article Role of ATP dependent potassium channels in myocardial ischaemia Garrett J Gross and John A Auchampach Recently, a class of potassium (K) channels has been discovered which are regulated by the intracellular level of ATP. These channels have been termed ATP dependent K channels (KATP)and have been found to exist in the heart, skeletal muscle, pancreatic p cells, brain, and smooth muscle. In this article, we discuss the function of the KAWchannel in the ischaemic myocardium and present evidence to suggest that activation of these channels may, on the one hand, result in a marked cardioprotective effect from reversible or irreversible electrical, functional or biochemical damage or, on the other hand, have the potential to produce electrical instability and a proarrhythmic effect. The therapeutic potential of potassium channel modulators is also discussed. Cardiovascular Research 1992;26: I01 1- 10 16 evidence that opening or blocking the KATPchannel may be beneficial or detrimental to the subsequent recovery of the heart after brief or prolonged periods of ischaemia followed by reperfusion, and that manipulation of this channel may have potential importance in the treatment of certain types of cardiovascular diseases.
Role of KATP channels in the coronary circulation Recent evidence suggests that KATP channels may be involved in the regulation of coronary blood flow, particularly during hypoxia and ischaemia. A role for KATP channels in hypoxic and ischaemic coronary vasodilatation is supported by the studies of Daut and coworker^'^ in which glibenclamide was shown to attenuate the reduction in coronary vascular resistance induced by hypoxia, ischaemia, metabolic inhibition, and adenosine in isolated guinea pig hearts. The effects of hypoxia were also mimicked by administration of cromakalim, an agent which opens KATP channels. Subsequently, Komaru et all5 showed that vasodilatation of small resistance arterioles (lo0 km) were not affected. Glibenclamide did not affect microvascular res onses to sodium nitroprusside. Furthermore, Aversano el a1I?found that reactive hyperaemia was markedly attenuated by intracoronary administration of glibenclamide in canine hearts. In addition, these investigatorsI6 showed that glibenclamide decreased the sensitivity of the coronary arteries to the dilator effect of intracoronary injections of adenosine but not to acetylcholine. In agreement, Belloni and Hintze17also found that glibenclamide attenuated the coronary vasodilator
Department of Pharmacology and Toxicology, Medical College of Wisconsin, 870 I Watertown Plank Road, Milwaukee, WI 53226, USA: G J Gross, J A Auchampach. Correspondence to Dr Gross.
Downloaded from by guest on May 19, 2016
S
ince the mid 1950s it has been well established that ischaemia results in a marked shortening of the cardiac action potential' which was attributed to activation of a time independent, outward potassium (K) current.However, the mechanisms or factors responsible for regulating this outward K current remained unknown until 1983, when Noma,3 using the patch clamp technique, reported the presence of K channels in guinea pig ventricular myocytes that were regulated primarily by the intracellular concentration of ATP and to a lesser degree by ADP, and he postulated that these channels may play a cardioprotective role during ischaemia or hypoxia. These channels were named ATP dependent K channels (4KATp) and have a l y bee! shown to exist in pancreatic p cells, skeletal muscle, brain, and vascular smooth m ~ s c l e . ~Subsequently other endogenous modulators of cellular KATPchannel activity have been identified and include the ADP/AT! ratio, certain nucleotide diphosphates, pH, and lactate. Unlike the pancreatic p cell where the KATPchannel has been shown to be intimately involved in insulin secretion: the function of the KATPchannel in the myocardium has not been well established. However, since the development of specific pharmacological modulators of this channel, namely the K channel openers which include nicorandil, cromakalim, pinacidil, aprikalim, and bimakalim,'" and the KATPchannel antagonists such as the sulphonylurea agents, glibenclamide and tolbutamide," and the non-sulphonylurea compound, sodium 5-hydrox decanoate,12 an ischaemia selective KATP channel blocker,i K substantial evidence has accumulated to suggest that opening these channels in the heart may play an important cardioprotective role, as assessed by functional recovery of regionally ischaemic or isolated globally ischaemic hearts or a reduction in infarct size in intact in situ hearts. Thus the major goal of the present article is to present
1012
Gross, Auchampach
res onses to adenosine in dogs. More recently, Samaha et all showed that infusion of glibenclamide into the coronary circulation of anaesthetised dogs and isolated perfused rabbit hearts resulted in an increase in coronary vascular resistance and evidence of tissue ischaemia. These findings led these authors18 to postulate that KATPchannels play an important role in regulating normal coronary vascular tone. Taken together, these results all suggest an important role for glibenclamide sensitive K channels in the regulation of coronary blood flow during ischaemia or hypoxia. However, there is still considerable debate as to whether these K channels found in the coronary circulation are similar to the KATPchannels identified in the heart.
P
Role of KATPchannels in myocardial ischaemia
Cardioprotective role of KATPchannels evidence
- in vitro
Recent studies in isolated globally ischaemic hearts and ventricular tissue suggest that the KATPchannel may serve a cardioprotective role during myocardial ischaemia. Grover et a13’ showed that several potassium channel openers, at concentrations which had no effect on contractile function under non-ischaemic conditions, improved functional recovery and preserved tissue adenine nucleotides following global ischaemia and reperfusion in isolated rat hearts and that pretreatment with glibenclamide3*or sodium 5-hydroxydecanoate13 prevented the beneficial effects of potassium channel openers, which suggests that these compounds are working via activation of KATPchannels. Furthermore, Cole et al” reported that preventing the opening of KATPchannels with glibenclamide worsened electrical and mechanical function following no flow ischaemia and reperfusion in isolated perfused guinea pig right ventricular walls, and enhancing the opening of KATP channels by pinacidil improved the recovery of electrical and mechanical activity. The effects of glibenclamide and pinacidil were concentration dependent and correlated directly with changes in action potential duration. Taken together, these studies support the original hypothesis of Noma3 that KATPchannels in the myocardium open during ischaemia and result in action potential shortening which serves to reduce the severity of injury, possibly by inhibiting calcium influx through voltage regulated calcium channels, thus preserving intracellular ATP. Other unknown metabolic effects, actions on different ionic channels or intracellular storage sites, or second messenger systems may also be involved in the cardioprotective effect that results from opening the KATP channel.
’’
Cardioprotective role of KATPchannels evidence Stunned myocardium
- in vivo
Brief intervals (5-20 minutes) of coronary artery occlusion insufficient to produce irreversible tissue damage result in prolonged abnormalities in regional contractile function, tissue blood flow, cellular ultrastructure and tissue adenine nucleotide content. This condition has been termed the stunned m y o c a r d i ~ m .Several ~~ mechanisms and mediators have been proposed as responsible for myocardial stunning; however, the most well established hypothesis suggests that postischaemic dysfunction is due to a disturbance in intracellular calcium homeostasis which may at least partially occur as a result of membrane damage produced by oxygen derived free radical^.^' The earliest studies to suggest a role for the KATPchannel in myocardial stunning were performed with the mixed potassium channel opener nitrate, nicorandil. Nicorandil was
Downloaded from by guest on May 19, 2016
Acute myocardial ischaemia is characterised by a rapid transition from normal electrical function to a state of metabolic, ionic, and electrical instability which results in a rapid decline in contractile function in the ischaemic area and is associated with a high incidence of arrhythmias.” Two pronounced features of acute myocardial ischaemia include action potential shortening and extracellular potassium accumulation, both of which occur within minutes of occlusion and have been attributed to a time independent outward potassium current.” While several theories have been proposed regarding the source of this increased potassium current, including ionic shifts in response to anion loss in the form of lactate, reduced sodium-potassium ATPase activity, and activation of calcium activated potassium channels,20recent studies suggest that opening of KAP channels is responsible, at least in part, for these alterations in potassium conductance. In hypoxic myocytes?’ globally ischaemic and regionally ischaemic hearts’. glibenclamide has been shown to reduce extracellular potassium accumulation, primarily during the early minutes of ischaemia, and to attenuate action potential shortening. However, the fact that glibenclamide was only partially effective in reducing K’ loss and preventing action potential shortening suggests that other factors may be important as sources of K’ loss during ischaemia and contribute to the shortening of action potential duration. Furthermore, potassium channels other than the KATP channel, such as those regulated by arachidonic acid or intracellular sodium, ma also be partly responsible for K+ loss during ischaemia! Thus, although there is good evidence to suggest that KAP channels open in response to ischaemia, other factors and channels may also be important. Since patch clamp studies reveal that KATPchannels are normally inybited by ATP at concentrations between 100 and 500 p,M‘ and it has been shown tha\eTP remains in the millimolar range during early ischaemia, the mechanism by which KAV channels open during ischaemia is unclear although several possibilities exist. First, it has been proposed that ATP is compartmentalised in myocardial cells and that certain pools which modulate KATPchannels may be selectively decreased during, ischaemia. In support of this hypothesis, Weiss and Lamp-’ reported that KATPchannels in cardiac myocytes are preferentially regulated by ATP which is produced by glycolysis, a metabolic pathway which is inhibited during hypoxia. A variation of this hypothesis is that a gradient may exist between the mitochondria and cell wall such that during ischaemia ATP at the site of the channel is reduced. A second possibility is that other intracellular factors which alter the sensitivity of the channel
to ATP are increased during ischaemia, such as ADP’* or hydrogen ions,” or that phosphorylation and/or G protein activity is increased by ischaemia or by factors that are released during ischaemia, such as adenosine. Finally; another hypothesis recently proposed by Findlay and Faivre suggests that less than 1% of available KATPchannels need to be activated to produce a 50% shortening of the cardiac action potential during ischaemia. Thus this “spare channel” concept may explain the apparent paradox in which minimal K+ loss occurs during ischaemia in spite of a substantial shortening of action potential duration.
Role of ATP dependent potassium channels in myocardial ischaemia
=
120
2”c
r-
\
*
*
*
*
T
T
p e
‘1
T
50
I
1013
T
+
T
I o Control 0
Q
2 Q
-80
-30
Occlusion ‘mnnI I -15
0 5 15
I
I
I 30
RP 52891 Occ
60
Time (min)
Figure 1 Percent segment shortening (%SS) of the ischaemicrepeQsed area before leji anterior descending coronary artery occlusion, during occlusion, and at various times during reperfusion in the vehicle treated group (Control), RP 52891 pretreated group (RP 52891 Occ), and the group treated with RP 52891 (aprikalim) only at reperfusion (RP 52891 Rep). All values are means, bars=SEM (n=B-IO each group). *pcO.O5 v Control.
Myocardial infa rction Prolonged coronary artery occlusion ultimately results in necrosis of ischaemic tissue or infarction unless timely blood flow is restored. The extent of infarction depends on the length of the occlusion period, oxygen demand during occlusion, area at risk, and the degree of collateral blood flow to the ischaemic area.4’ There is also a considerable body of evidence implicating the inflammatory response and the neutrophil in an evolving myocardial infar~tion.~’ The effects of potassium channel openers on myocardial infarct size are equivocal. Pinacidil has been shown to have no effect4“ or to increase infarct size in conscious dogs.@ However in both of these studies pinacidil was administered at a dose which produced marked hypotension, reflex tachycardia, and a decrease in collateral blood flow, so it is not surprising that infarct size was not changed or increased. In contrast, Grover and colleagues3’found that when cromakalim and pinacidil were given by intracoronary injection to dogs at doses that produced no haemodynamic benefit, a marked reduction in myocardial infarct size was observed. In agreement, Auchampach et a14s also found that systemic
IF/AAR
IFlLV
Figure 2 Effects of RP 52891 (aprikalim) on the area at risk (AAR)expressed as a percent of the leji ventricular (LV) weight and infarct size ( I F ) as a percent of the AAR or LK Columns are means, bars=SEM (n=9 each group). *p
1011
Review article Role of ATP dependent potassium channels in myocardial ischaemia Garrett J Gross and John A Auchampach Recently, a class of potassium (K) channels has been discovered which are regulated by the intracellular level of ATP. These channels have been termed ATP dependent K channels (KATP)and have been found to exist in the heart, skeletal muscle, pancreatic p cells, brain, and smooth muscle. In this article, we discuss the function of the KAWchannel in the ischaemic myocardium and present evidence to suggest that activation of these channels may, on the one hand, result in a marked cardioprotective effect from reversible or irreversible electrical, functional or biochemical damage or, on the other hand, have the potential to produce electrical instability and a proarrhythmic effect. The therapeutic potential of potassium channel modulators is also discussed. Cardiovascular Research 1992;26: I01 1- 10 16 evidence that opening or blocking the KATPchannel may be beneficial or detrimental to the subsequent recovery of the heart after brief or prolonged periods of ischaemia followed by reperfusion, and that manipulation of this channel may have potential importance in the treatment of certain types of cardiovascular diseases.
Role of KATP channels in the coronary circulation Recent evidence suggests that KATP channels may be involved in the regulation of coronary blood flow, particularly during hypoxia and ischaemia. A role for KATP channels in hypoxic and ischaemic coronary vasodilatation is supported by the studies of Daut and coworker^'^ in which glibenclamide was shown to attenuate the reduction in coronary vascular resistance induced by hypoxia, ischaemia, metabolic inhibition, and adenosine in isolated guinea pig hearts. The effects of hypoxia were also mimicked by administration of cromakalim, an agent which opens KATP channels. Subsequently, Komaru et all5 showed that vasodilatation of small resistance arterioles (lo0 km) were not affected. Glibenclamide did not affect microvascular res onses to sodium nitroprusside. Furthermore, Aversano el a1I?found that reactive hyperaemia was markedly attenuated by intracoronary administration of glibenclamide in canine hearts. In addition, these investigatorsI6 showed that glibenclamide decreased the sensitivity of the coronary arteries to the dilator effect of intracoronary injections of adenosine but not to acetylcholine. In agreement, Belloni and Hintze17also found that glibenclamide attenuated the coronary vasodilator
Department of Pharmacology and Toxicology, Medical College of Wisconsin, 870 I Watertown Plank Road, Milwaukee, WI 53226, USA: G J Gross, J A Auchampach. Correspondence to Dr Gross.
Downloaded from by guest on May 19, 2016
S
ince the mid 1950s it has been well established that ischaemia results in a marked shortening of the cardiac action potential' which was attributed to activation of a time independent, outward potassium (K) current.However, the mechanisms or factors responsible for regulating this outward K current remained unknown until 1983, when Noma,3 using the patch clamp technique, reported the presence of K channels in guinea pig ventricular myocytes that were regulated primarily by the intracellular concentration of ATP and to a lesser degree by ADP, and he postulated that these channels may play a cardioprotective role during ischaemia or hypoxia. These channels were named ATP dependent K channels (4KATp) and have a l y bee! shown to exist in pancreatic p cells, skeletal muscle, brain, and vascular smooth m ~ s c l e . ~Subsequently other endogenous modulators of cellular KATPchannel activity have been identified and include the ADP/AT! ratio, certain nucleotide diphosphates, pH, and lactate. Unlike the pancreatic p cell where the KATPchannel has been shown to be intimately involved in insulin secretion: the function of the KATPchannel in the myocardium has not been well established. However, since the development of specific pharmacological modulators of this channel, namely the K channel openers which include nicorandil, cromakalim, pinacidil, aprikalim, and bimakalim,'" and the KATPchannel antagonists such as the sulphonylurea agents, glibenclamide and tolbutamide," and the non-sulphonylurea compound, sodium 5-hydrox decanoate,12 an ischaemia selective KATP channel blocker,i K substantial evidence has accumulated to suggest that opening these channels in the heart may play an important cardioprotective role, as assessed by functional recovery of regionally ischaemic or isolated globally ischaemic hearts or a reduction in infarct size in intact in situ hearts. Thus the major goal of the present article is to present
1012
Gross, Auchampach
res onses to adenosine in dogs. More recently, Samaha et all showed that infusion of glibenclamide into the coronary circulation of anaesthetised dogs and isolated perfused rabbit hearts resulted in an increase in coronary vascular resistance and evidence of tissue ischaemia. These findings led these authors18 to postulate that KATPchannels play an important role in regulating normal coronary vascular tone. Taken together, these results all suggest an important role for glibenclamide sensitive K channels in the regulation of coronary blood flow during ischaemia or hypoxia. However, there is still considerable debate as to whether these K channels found in the coronary circulation are similar to the KATPchannels identified in the heart.
P
Role of KATPchannels in myocardial ischaemia
Cardioprotective role of KATPchannels evidence
- in vitro
Recent studies in isolated globally ischaemic hearts and ventricular tissue suggest that the KATPchannel may serve a cardioprotective role during myocardial ischaemia. Grover et a13’ showed that several potassium channel openers, at concentrations which had no effect on contractile function under non-ischaemic conditions, improved functional recovery and preserved tissue adenine nucleotides following global ischaemia and reperfusion in isolated rat hearts and that pretreatment with glibenclamide3*or sodium 5-hydroxydecanoate13 prevented the beneficial effects of potassium channel openers, which suggests that these compounds are working via activation of KATPchannels. Furthermore, Cole et al” reported that preventing the opening of KATPchannels with glibenclamide worsened electrical and mechanical function following no flow ischaemia and reperfusion in isolated perfused guinea pig right ventricular walls, and enhancing the opening of KATP channels by pinacidil improved the recovery of electrical and mechanical activity. The effects of glibenclamide and pinacidil were concentration dependent and correlated directly with changes in action potential duration. Taken together, these studies support the original hypothesis of Noma3 that KATPchannels in the myocardium open during ischaemia and result in action potential shortening which serves to reduce the severity of injury, possibly by inhibiting calcium influx through voltage regulated calcium channels, thus preserving intracellular ATP. Other unknown metabolic effects, actions on different ionic channels or intracellular storage sites, or second messenger systems may also be involved in the cardioprotective effect that results from opening the KATP channel.
’’
Cardioprotective role of KATPchannels evidence Stunned myocardium
- in vivo
Brief intervals (5-20 minutes) of coronary artery occlusion insufficient to produce irreversible tissue damage result in prolonged abnormalities in regional contractile function, tissue blood flow, cellular ultrastructure and tissue adenine nucleotide content. This condition has been termed the stunned m y o c a r d i ~ m .Several ~~ mechanisms and mediators have been proposed as responsible for myocardial stunning; however, the most well established hypothesis suggests that postischaemic dysfunction is due to a disturbance in intracellular calcium homeostasis which may at least partially occur as a result of membrane damage produced by oxygen derived free radical^.^' The earliest studies to suggest a role for the KATPchannel in myocardial stunning were performed with the mixed potassium channel opener nitrate, nicorandil. Nicorandil was
Downloaded from by guest on May 19, 2016
Acute myocardial ischaemia is characterised by a rapid transition from normal electrical function to a state of metabolic, ionic, and electrical instability which results in a rapid decline in contractile function in the ischaemic area and is associated with a high incidence of arrhythmias.” Two pronounced features of acute myocardial ischaemia include action potential shortening and extracellular potassium accumulation, both of which occur within minutes of occlusion and have been attributed to a time independent outward potassium current.” While several theories have been proposed regarding the source of this increased potassium current, including ionic shifts in response to anion loss in the form of lactate, reduced sodium-potassium ATPase activity, and activation of calcium activated potassium channels,20recent studies suggest that opening of KAP channels is responsible, at least in part, for these alterations in potassium conductance. In hypoxic myocytes?’ globally ischaemic and regionally ischaemic hearts’. glibenclamide has been shown to reduce extracellular potassium accumulation, primarily during the early minutes of ischaemia, and to attenuate action potential shortening. However, the fact that glibenclamide was only partially effective in reducing K’ loss and preventing action potential shortening suggests that other factors may be important as sources of K’ loss during ischaemia and contribute to the shortening of action potential duration. Furthermore, potassium channels other than the KATP channel, such as those regulated by arachidonic acid or intracellular sodium, ma also be partly responsible for K+ loss during ischaemia! Thus, although there is good evidence to suggest that KAP channels open in response to ischaemia, other factors and channels may also be important. Since patch clamp studies reveal that KATPchannels are normally inybited by ATP at concentrations between 100 and 500 p,M‘ and it has been shown tha\eTP remains in the millimolar range during early ischaemia, the mechanism by which KAV channels open during ischaemia is unclear although several possibilities exist. First, it has been proposed that ATP is compartmentalised in myocardial cells and that certain pools which modulate KATPchannels may be selectively decreased during, ischaemia. In support of this hypothesis, Weiss and Lamp-’ reported that KATPchannels in cardiac myocytes are preferentially regulated by ATP which is produced by glycolysis, a metabolic pathway which is inhibited during hypoxia. A variation of this hypothesis is that a gradient may exist between the mitochondria and cell wall such that during ischaemia ATP at the site of the channel is reduced. A second possibility is that other intracellular factors which alter the sensitivity of the channel
to ATP are increased during ischaemia, such as ADP’* or hydrogen ions,” or that phosphorylation and/or G protein activity is increased by ischaemia or by factors that are released during ischaemia, such as adenosine. Finally; another hypothesis recently proposed by Findlay and Faivre suggests that less than 1% of available KATPchannels need to be activated to produce a 50% shortening of the cardiac action potential during ischaemia. Thus this “spare channel” concept may explain the apparent paradox in which minimal K+ loss occurs during ischaemia in spite of a substantial shortening of action potential duration.
Role of ATP dependent potassium channels in myocardial ischaemia
=
120
2”c
r-
\
*
*
*
*
T
T
p e
‘1
T
50
I
1013
T
+
T
I o Control 0
Q
2 Q
-80
-30
Occlusion ‘mnnI I -15
0 5 15
I
I
I 30
RP 52891 Occ
60
Time (min)
Figure 1 Percent segment shortening (%SS) of the ischaemicrepeQsed area before leji anterior descending coronary artery occlusion, during occlusion, and at various times during reperfusion in the vehicle treated group (Control), RP 52891 pretreated group (RP 52891 Occ), and the group treated with RP 52891 (aprikalim) only at reperfusion (RP 52891 Rep). All values are means, bars=SEM (n=B-IO each group). *pcO.O5 v Control.
Myocardial infa rction Prolonged coronary artery occlusion ultimately results in necrosis of ischaemic tissue or infarction unless timely blood flow is restored. The extent of infarction depends on the length of the occlusion period, oxygen demand during occlusion, area at risk, and the degree of collateral blood flow to the ischaemic area.4’ There is also a considerable body of evidence implicating the inflammatory response and the neutrophil in an evolving myocardial infar~tion.~’ The effects of potassium channel openers on myocardial infarct size are equivocal. Pinacidil has been shown to have no effect4“ or to increase infarct size in conscious dogs.@ However in both of these studies pinacidil was administered at a dose which produced marked hypotension, reflex tachycardia, and a decrease in collateral blood flow, so it is not surprising that infarct size was not changed or increased. In contrast, Grover and colleagues3’found that when cromakalim and pinacidil were given by intracoronary injection to dogs at doses that produced no haemodynamic benefit, a marked reduction in myocardial infarct size was observed. In agreement, Auchampach et a14s also found that systemic
IF/AAR
IFlLV
Figure 2 Effects of RP 52891 (aprikalim) on the area at risk (AAR)expressed as a percent of the leji ventricular (LV) weight and infarct size ( I F ) as a percent of the AAR or LK Columns are means, bars=SEM (n=9 each group). *p
