K⁺ channels in the coronary microvasculature of the ischemic heart

Abstract

Ischemic heart disease is the leading cause of death and a major public health and economic burden worldwide with expectations of predicted growth in the foreseeable future. It is now recognized clinically that flow-limiting stenosis of the large coronary conduit arteries as well as microvascular dysfunction in the absence of severe stenosis can each contribute to the etiology of ischemic heart disease. The primary site of coronary vascular resistance, and control of subsequent coronary blood flow, is found in the coronary microvasculature, where small changes in radius can have profound impacts on myocardial perfusion. Basal active tone and responses to vasodilators and vasoconstrictors are paramount in the regulation of coronary blood flow and adaptations in signaling associated with ion channels are a major factor in determining alterations in vascular resistance and thereby myocardial blood flow. K⁺ channels are of particular importance as contributors to all aspects of the regulation of arteriole resistance and control of perfusion into the myocardium because these channels dictate membrane potential, the resultant activity of voltage-gated calcium channels, and thereby, the contractile state of smooth muscle. Evidence also suggests that K⁺ channels play a significant role in adaptations with cardiovascular disease states. In this review, we highlight our research examining the role of K⁺ channels in ischemic heart disease and adaptations with exercise training as treatment, as well as how our findings have contributed to this area of study.

1. Introduction

Cardiovascular disease is the leading cause of death worldwide with ischemic heart disease persisting as the most prevalent form (GBD 2015 Disease and Injury Incidence and Prevalence Collaborators, 2016). In physiological settings, the coronary circulation supplies the myocardium with blood based upon the metabolic demands of the heart (Duncker, Koller, Merkus, & Canty, 2015). Because oxygen extraction is high under resting conditions, increases in metabolic activity, such as during exercise, are met primarily by vasodilation of the coronary vasculature to increase coronary blood flow proportionally to the demands of the myocardium. Pathophysiological conditions, such as ischemia, where there is inadequate blood flow to the myocardium, can be brought about by the build-up and rupture of atherosclerotic plaques, severe occlusion of an epicardial conduit artery, or microvascular dysfunction, resulting in a mismatch between metabolic needs and perfusion of the myocardium. Oftentimes, patients that experience angina yet have no angiographic evidence for significant obstructive coronary lesions or macrovascular (conduit artery) spasm, display functional and structural impairments of the coronary microcirculation which lead to impaired vasodilation and thereby, myocardial ischemia (Lee et al., 2022). Alternatively, coronary microvascular spasm can also occur, resulting in significant reductions in myocardial blood flow. Mechanisms that contribute to coronary microvascular dysfunction may include an imbalance between vasodilators and vasoconstrictors released by the endothelium or enhanced tone of the vascular smooth muscle that may occur independently of endothelium-derived vasodilators (Vancheri, Longo, Vancheri, & Henein, 2020).

Improvements in perfusion of the ischemic myocardium can be brought about by increases in collateral development as well as enhanced vasodilation responses of the innate microvasculature in the ischemic regions of the heart. There is substantial evidence that exercise training is a valuable, non-pharmacological therapeutic option for cardiovascular disease (Duncker & Bache, 2008; Griffin, Laughlin, & Parker, 1999; Griffin, Woodman, Price, Laughlin, & Parker, 2001). A significant body of literature has supported a role for physical activity in patients with coronary artery disease to induce improvements in coronary endothelial function and endothelium-dependent relaxation in the diseased coronary vasculature. Early work by Hambrecht and colleagues (Hambrecht et al., 2000) revealed that after completion of a 4-week cycle ergometry exercise regimen, atherosclerotic coronary disease patients displayed improved endothelium-dependent vasodilation in conduit coronary arteries and arterioles and enhanced coronary flow reserve. Considerable evidence over the years has confirmed these data and repeatedly shown enhanced vasodilation capacity in diseased patients and animal models after exercise training (Conraads et al., 2015; Koller et al., 2022). In contrast, direct evidence of increased collateralization with exercise training is limited in human patients with coronary artery disease, although animal models support a role for collateral expansion with exercise (Bloor, White, & Sanders, 1984; Cohen, Yipintsoi, & Scheuer, 1982; Roth et al., 1990). Numerous studies have reported no increase in collateralization after exercise training of patients (Merkus, Muller-Delp, & Heaps, 2021). Only a small number of studies have provided direct angiographic evidence that exercise training potentiated coronary collateralization along with improved perfusion by collateral vessels and improvements in myocardial contractility in diseased patients (Belardinelli, Belardinelli, & Shryock, 2001; Belardinelli, Georgiou, Ginzton, Cianci, & Purcaro, 1998). However, because assessment of collateralization by angiography has significant limitations in detection of vessels smaller than ~200 μm, the assessment of collateral development is insufficient in humans. Thus, despite exercise training-induced improvements in contractility of myocardial regions compromised by ischemia, the relative contributions of collateralization and enhanced reactivity of the microvasculature supplying that region remain unclear.

Coronary vascular tone is highly regulated in order to match oxygen delivery with metabolic demand. Vascular smooth muscle cells comprise the majority of the vascular wall and dictate blood flow by changes in contractile state and thus lumen diameter. Indeed, consistent with Pouseille’s Law, small changes in vascular radius markedly impact blood flow into the dependent region. Vascular smooth muscle cells express a variety of ion channels that impact cellular membrane potential, with K⁺ channels having the largest impact (Tykocki, Boerman, & Jackson, 2017). There is a vast number of K⁺ channel classes, each with numerous isoforms, including voltage-gated K⁺ (Kv) channels, calcium-activated K⁺ (KCa) channels, adenosine triphosphate-sensitive K⁺ (KATP) channels, and inward-rectifying K⁺ (KIR) channels (Fig. 1). Numerous studies have identified members of each class within the coronary smooth muscle, most notably, Kv, KATP, KIR, and large-conductance, calcium-dependent KCa (BKCa) channels (Tykocki et al., 2017).

Fig. 1.

Potassium channel pathways in the vascular smooth muscle and endothelium. Schematic depicts depolarization inducing activation of voltage-gated calcium channels (VGCC) and voltage-gated potassium (Kv) channels as denoted by arrows. Activation of VGCC allows for calcium movement into the cell, stimulating contraction. Large-conductance calcium-activated potassium (BKCa) channels open in response to increased intracellular calcium, allowing for K⁺ efflux to counteract the depolarization and contractile response of calcium influx. Kv channel activation via depolarization leads to K⁺ efflux. Together, Kv and BKCa channels allow K⁺ efflux to hyperpolarize the cell membrane, combating VGCC stimulation, inducing relaxation of the contractile machinery. Intracellular calcium is also transported into the mitochondria via mitochondrial calcium uniporters as depicted in the smooth muscle cell. The reactive oxygen species (ROS) produced by the mitochondria can both stimulate and inhibit the activity of the Kv and BKCa channels. Within the smooth muscle cell are PKA and PKG proteins, which can be activated by PGI2 and NO respectively, that have been found to stimulate numerous potassium channels. In addition to PKA and PKG, the KATP channel is dependent on an increased ADP:ATP ratio for stimulation, while decreased ADP:ATP leads to inhibition. The stimulated KATP channel allows for outward potassium transport, resulting in inhibition of contraction. On the endothelial cell, there are the intermediate and small conductance calcium-activated potassium channels as well as KATP channels. Depolarization inhibits inward-rectifying potassium channel (KIR) as indicated with the dashed blunt arrow. Endothelial cell mitochondria can produce ROS (e.g., superoxide) which is then converted to hydrogen peroxide via superoxide dismutase (SOD). Hydrogen peroxide can activate various potassium channels. Under ischemic conditions, it has been proposed that numerous potassium channels including Kv, BKCa, and KATP, may act as redox or oxygen sensors, involved in regulation of coronary blood flow. See text for further details and references. Created with Biorender.com.

Smooth muscle resting membrane potential is largely controlled by K⁺ leak channels on the membrane surface with Kv and BKCa channels also playing a role (Gutterman, Miura, & Liu, 2005). Depolarization of the smooth muscle plasma membrane leads to activation of Kv channels and subsequent hyperpolarization. This reduces the activity of voltage-gated Ca²⁺ channels, diminishing intracellular Ca²⁺ concentrations, resulting in smooth muscle relaxation and vasodilation (Kim, Appel, Vetterkind, Gangopadhyay, & Morgan, 2008). In parallel, increased intracellular Ca²⁺ concentrations, resulting from membrane depolarization, activate KCa channels yielding membrane hyperpolarization, diminishing intracellular Ca²⁺ concentrations via closure of voltage-gated Ca²⁺ channels, and subsequent vasodilation (Berkefeld, Fakler, & Schulte, 2010; Hill, Yang, Ella, Davis, & Braun, 2010). Due to the high membrane resistance of vascular smooth muscle cells, small changes in K⁺ channel activity can have significant impact on arteriolar tone (Nelson, Patlak, Worley, & Standen, 1990). In addition, numerous K⁺ channels including Kv channels of the Kv1.5 and Kv2.1 subtype (Dwenger, Ohanyan, Navedo, & Nystoriak, 2018; Nishijima et al., 2017; Ohanyan et al., 2015; Rogers et al., 2006; Rogers, Chilian, Bratz, Bryan, & Dick, 2007; Shimizu, Yokoshiki, Sperelakis, & Paul, 2000; Thorne, Conforti, & Paul, 2002), BKCa (Gebremedhin et al., 1994; López-Barneo, del Toro, Levitsky, Chiara, & Ortega-Sáenz, 2004; Nelson & Quayle, 1995), and KATP (Dart & Standen, 1995; Kamekura et al., 1999; Lee, Kim, & Kang, 1998) channels have been proposed as redox or O₂ sensors and involved in coronary vasodilation associated with ischemic conditions. With each of these subfamilies possessing multiple isoforms, identifying which K⁺ channels are involved in ischemia/hypoxia-induced vasodilation is complex. Further, ion channel expression and regulation are different in macro- vs. microvasculature as shown by greater frequency of spontaneous transient outward currents (BKCa) (Mokelke, Dietz, Eckman, Nelson, & Sturek, 2005) and voltage-gated Ca²⁺ channels (Bowles, Hu, Laughlin, & Sturek, 1998) in microvascular compared to macrovascular smooth muscle. These differences point to the specificity of coronary microvascular regulation. Thus, exploration of the contribution of specific K⁺ channel isoforms is actively being pursued in microvascular smooth muscle and much insight remains to be gained from future study of the role of specific channel isoforms in coronary blood flow under control and compromised conditions as well as adaptations associated with exercise and other therapies.

Another family of potassium channels that have an effect on vascular tone are the inward-rectifying potassium (KIR) channels, expressed in both smooth muscle and endothelial cells (Hibino et al., 2010; Quayle, Nelson, & Standen, 1997). Under physiological conditions, the inward-rectifying potassium channel (KIR2.1) regulates vasodilation via nitric oxide production (Ahn et al., 2017; Fancher et al., 2018). Nelson and colleagues (Sonkusare, Dalsgaard, Bonev, & Nelson, 2016) determined that endothelial cell KIR2 channels may amplify the vasodilation signals of resistance arteries produced by SKCa and IKCa channels. Others have shown a small role for endothelial cell KIR channels in dilation of human coronary arterioles in response to slight elevations in extracellular potassium levels (Miura, Toyama, Pratt, & Gutterman, 2011), such as those observed during exercise (Wilson, Kapoor, & Krishna, 1994). The role of these channels in patients and animal models of ischemic heart disease have not been explored in detail.

The focus of this review is to highlight our research examining the role of microvascular smooth muscle K⁺ channels in ischemic heart disease and how our research has contributed to the field of study. We will focus primarily on adaptations in the coronary microcirculation distal to epicardial stenosis/occlusion as a key mechanism regulating coronary ischemia.

2. Porcine model of ischemic heart disease

Our laboratory has had a long-standing interest in the mechanisms by which exercise training drives adaptations in the microcirculation of the ischemic heart that increase blood flow into compromised myocardium and thereby, improve the function (e.g., contractility) of the heart. Indeed, the regulation of blood flow through the coronary microvasculature determines function of not only the heart but the entire body, since inadequate myocardial blood flow negatively impacts the ability of the heart to pump effectively, limiting delivery of blood to the periphery. Over the years, we have examined exercise training-induced adaptations of the coronary microcirculation using porcine models of both hypercholesterolemia and chronic coronary artery occlusion.

We use a porcine model of chronic coronary artery occlusion and exercise training to gain a better understanding of the fundamental cellular and molecular mechanisms that underlie exercise-induced cardioprotection. Our goal is to provide insight into new therapeutic targets for increasing blood flow into compromised myocardium. For much of the work we will discuss in this review, we have used sexually-mature Yucatan miniature swine (6–7 months of age) surgically instrumented with an ameroid constrictor around the proximal left circumflex coronary artery. Animals are placed under general anesthesia (2–3% isoflurane and supplemental O₂) during aseptic surgery and a left lateral thoracotomy is performed in the fourth intercostal space. The underlying pericardium is opened to expose the proximal left circumflex artery, which is then dissected free of surrounding tissue and an ameroid constrictor (2.5- to 3.5-mm inner diameter) is placed around the artery (Fig. 2A). The diameter of the occluder is selected upon visual inspection of the artery to initially provide a secure but non-constrictive fit. The pericardium is then closed and the thoracotomy is repaired in tissue layers. Pigs are closely monitored during surgery and surgical recovery, receiving pain medication as needed. Ameroid closure and total occlusion ensues approximately 3 weeks after ameroid placement (White, Carroll, Magnet, & Bloor, 1992). Pigs undergo postoperative recovery for 8 weeks before the sedentary or exercise training experimental protocol is initiated. Previous studies have demonstrated that between 3 and 7 weeks following ameroid placement, the development of collateral vessels in response to occlusion restores cardiac function under resting conditions and further collateral development is not observed (Roth et al., 1987). Thus, collateral vessel development in response to exercise can be measured independent of surgically-induced collateral growth. The schematic in Fig. 2B shows the placement of the ameroid occluder around the proximal left circumflex artery and the consequent collateral-dependent region produced after complete occlusion, as well as the nonoccluded region served by the left anterior descending artery.

Fig. 2.

(A) During aseptic surgery, a left lateral thoracotomy is performed in the fourth intercostal space. Finochietto retractor (pediatric) is used to maintain a clear surgical field. The underlying pericardium is opened to expose the myocardial region containing the proximal LCX artery. The LCX artery is dissected free of surrounding tissue and an ameroid constrictor (stainless steel device) is placed around the artery. Pericardium is held in the open position during placement of the ameroid using silk suture (2-0). Gauze is used beneath the silk suture to protect underlying tissues. (B) 3 weeks post-operatively, complete occlusion of the LCX artery is achieved, generating a downstream, collateral-dependent myocardium. This tissue is now reliant on expansion of the collateral circulation supported by flow from both the nonoccluded left anterior descending (LAD) and right coronary (RCA) arteries. The myocardium supplied by the LAD is defined as the nonoccluded control region.

This porcine model is widely recognized as highly clinically relevant because it replicates many human adaptations to both ischemic heart disease and exercise training (Cohen, 1985; Douglas, 1972; O’Konski, White, Longhurst, Roth, & Bloor, 1986; Roth et al., 1987, 1990; Schaper, 1971a; White et al., 1992; White & Bloor, 1981, 1992; White, Roth, & Bloor, 1986). The cardiovascular system and in particular the coronary circulation of the pig has many anatomic and physiologic attributes that mimic those of humans (Douglas, 1972; Schaper, 1971a; White & Bloor, 1981). Furthermore, the pig is an excellent model of progressive coronary artery occlusion because development of collateral vessels in the pig heart is sufficient to provide normal blood flow to the collateral-dependent myocardium at rest; however, blood flow during stress (e.g., exercise) remains compromised and unable to support regional myocardial function (O’Konski et al., 1986; Roth et al., 1987, 1990; White et al., 1986, 1992). Similarly, the human heart demonstrates few innate collateral vessels and growth of new vessels typically occurs as an extensive network of functionally significant collaterals in the endocardial and mid-myocardial layers (Cohen, 1985; White & Bloor, 1992; White et al., 1992). Furthermore, coronary artery disease patient populations often lack clinical symptoms at rest, but exhibit significant signs of persistent regional myocardial ischemia and contractile dysfunction (angina, ECG abnormalities) during periods of increased cardiac workload (Kolibash et al., 1982; Niebauer et al., 1995; Schuler et al., 1992) similar to that observed in pigs (O’Konski et al., 1986; Roth et al., 1987; White et al., 1992). Finally, histological analyses demonstrate that this porcine model displays an average infarct size of only ~1% of the left ventricular myocardium (Roth et al., 1987, 1990), thus, this model is an effective model of ischemic heart disease.

In contrast to humans and pigs, the canine heart possesses an abundance of innate collateral vessels, which can be recruited rapidly under conditions of ischemia (Cohen, 1985; Schaper, 1971b). Furthermore, collateralization of the canine heart generally occurs as an epicardial collateral network in contrast to the endocardial distribution in humans and pigs. The combination of vessel recruitment and the growth of new collaterals are sufficient to restore adequate blood flow to the compromised myocardium during rest and often under conditions of increased myocardial oxygen demand (Bache & Schwartz, 1983; Cohen & Yipintsoi, 1981; Eckstein, 1957; Lambert, Hess, & Bache, 1977; Longhurst, Motohara, Atkins, & Ordway, 1985). Thus, study of vascular adaptations to ischemia in the canine model may not be highly translatable to humans because the dog does not experience the level of ischemia found with obstructive coronary artery disease observed in human patients. Alternatively, the rapid collateral vessel growth observed in the dog may serve as an advantage in elucidating mechanisms of collateral development that can be exploited therapeutically to improve collateral development in humans. Rodents (rats and mice) are clearly inadequate for coronary microvascular responses to occlusion and exercise because the high resting heart rate indicates very high metabolic rate and the minimal changes in heart rate during exercise are vastly different than that observed in humans (Duncker & Bache, 2008).

In addition to similarities in response to occlusive artery disease, the porcine model also mimics responses to exercise training in that pigs are able to increase their maximal oxygen consumption with persistent exercise training and the maximal capacity of the coronary circulation is similar to that of humans (White et al., 1986). Furthermore, analogous to that observed in humans, swine demonstrate redistribution of blood flow from visceral tissue to skeletal muscle during exercise and display similar maximal oxygen consumption during exercise (Armstrong, Delp, Goljan, & Laughlin, 1987). Furthermore, despite perfusion deficits during increased cardiac workload (Roth et al., 1987), persistent exercise training has been shown to improve the perfusion deficit and contractile dysfunction of the myocardium at risk (Roth et al., 1990).

To confirm previous reports (O’Konski et al., 1986; Roth et al., 1987, 1990; White et al., 1986; White & Bloor, 1992), we performed dobutamine stress echocardiography to examine regional myocardial wall function in our chronically occluded swine using peak systolic myocardial velocity as measure of function (Fig. 3). Our data indicate that under resting conditions, myocardial function is not impaired in the collateral-dependent region compared with the nonoccluded region. In contrast, at high dobutamine doses (20 g/kg/min), when heart rates of ~190 bpm are attained, function of the collateral-dependent region is markedly diminished compared with the nonoccluded region. This finding is of great significance because during each bout of treadmill exercise, our exercise-trained animals attain heart rates of ~190–200 bpm (~65–70% of maximal heart rate), suggesting that during exercise, our pigs likely experience ischemic episodes that may contribute to adaptations of the collateral-dependent region. Completion of such studies under cardiac stress are important since patients with advanced coronary artery disease often lack clinical symptoms at rest, but exhibit significant signs of myocardial ischemia (angina, ECG abnormalities) during periods of increased cardiac workload. Taken together, these data indicate that the porcine model of progressive, chronic occlusion is a representative and valid model of stress-induced ischemic heart disease that closely mimics human ischemic heart disease.

Fig. 3.

Regional myocardial wall function as determined by dobutamine stress echocardiography using peak systolic myocardial velocity as measure of function. Wall function was determined in both the myocardial region supplied by the left anterior descending artery (nonoccluded) and the collateral-dependent region previously supplied by the left circumflex coronary artery. At rest (0 μg/kg/min dobutamine), peak velocity is similar between the nonoccluded and collateral-dependent regions, whereas during increased demand (20 μg/kg/min dobutamine), peak velocity is markedly reduced in the collateral-dependent compared with the nonoccluded region.

Serial coronary angiographic images (Fig. 4) taken 21-weeks following placement of the ameroid constrictor confirm that blood flow through the left circumflex artery at the site of occlusion was completely abolished by the ameroid constrictor. Relatively large, tortuous collateral vessels supplying the left circumflex artery distal to occlusion are clearly visible in both panels of the angiogram. These collaterals appear to originate from the proximal left anterior descending and the left circumflex artery (proximal to occluder) and “bypass” the occlusion. Smaller intramyocardial collaterals are likely also present (White et al., 1992) but are difficult to visualize by angiography (Heaps & Parker, 2011). Importantly, because complete ameroid-induced left circumflex artery occlusion is evident, the myocardial region formerly perfused by the native left circumflex artery is, by definition, now fully dependent upon perfusion via collateral vessels such as those observed in Fig. 4.

Fig. 4.

Sequential angiograms of the porcine model of progressive chronic coronary artery occlusion and collateral development. 22 weeks after surgical placement of an ameroid occluder around the proximal left circumflex coronary artery, the pig was placed under general anesthesia and hemodynamics were monitored throughout the procedure. Arterial access was obtained by surgical cutdown of the carotid artery. The left main coronary artery was catheterized with a 6F guiding catheter (Vista Brite Tip; Cordis) introduced over a 0.035-in. guidewire. Selective coronary angiography was performed with nonionic contrast agent (Oxilan; 350 mgL/mL). (A–C): regionally enhanced serial images emphasizing the cardiac silhouette to provide additional detail of coronary vasculature. White arrowheads identify collateral vessels supplying the left circumflex artery (LCX) distal to occlusion. The left anterior descending (LAD) artery demonstrates unobstructed flow. LMC, left main coronary artery catheter; PVC, pulmonary vein catheter. Adapted from Heaps, C. L., & Parker, J. L. (2011). Effects of exercise training on coronary collateralization and control of collateral resistance. Journal of Applied Physiology, 111(2), 587–598.

3. Role of K⁺ channels in basal active tone of arterioles from ischemic myocardium

Our early work evaluating the coronary microcirculation in ischemic heart disease explored the effect of exercise training on K⁺ channels as mediators of basal active tone in the microvasculature distal to chronic occlusion. These studies revealed that 4-aminopyridine-sensitive Kv channels contribute markedly to resting tone in arterioles of nonoccluded and collateral-dependent myocardial regions in both sedentary and exercise-trained animals. Intriguingly, the contractile response to Kv channel blockade was statistically greater in arterioles from the collateral-dependent region of exercise-trained pigs, indicating that Kv channels contribute to the regulation of basal tone to a greater extent in arterioles distal to occlusion after exercise training (Heaps, Mattox, Kelly, Meininger, & Parker, 2006). Kv channels that have been shown to be 4-aminopyridine-sensitive include primarily the Kv1-Kv4 subfamilies (Tykocki et al., 2017), suggesting that some members of these channel subtypes contribute significantly to basal active tone in the coronary circulation with marked induction in the collateral-dependent region after exercise training. On the other hand, BKCa channel contribution to basal tone was negligible in all vessel treatment groups (Heaps et al., 2006), a finding that was also demonstrated in large conduit arteries from these animals (Deer & Heaps, 2013). In a previous study, Bowles, Laughlin, and Sturek (1998) demonstrated an increased role for both 4-AP-sensitive Kv channels and BKCa channels in the maintenance of basal tone after exercise training in porcine epicardial conduit artery segments, although the role of 4-AP-sensitive Kv channels was markedly greater than that of BKCa channels.

Paradoxically, in addition to an increased contribution of K⁺ channels to basal tone, we also observed enhanced Ca²⁺-dependent basal active tone and increased nitric oxide contributions to resting tone (Heaps et al., 2006). Taken together, these data suggest that exercise training drives a parallel upregulation of seemingly contradictory signaling mechanisms that may ultimately function to optimize the control of coronary basal tone. Adaptations in both vasodilation and vasoconstriction pathways may more precisely maintain appropriate blood flow to meet the metabolic demands of the at-risk myocardium, as well as contribute to the maintenance of coronary flow reserve. The stimulus for these adaptations may be most marked in the collateral-dependent region because of coupling of intermittent ischemia with mechanical forces on the vascular wall during each bout of exercise. Indeed, exercise-induced myocardial ischemia persists in this animal model despite improvements in coronary blood flow and myocardial contractile function after chronic exercise training (Roth et al., 1990).

4. Kv channels in arteriolar smooth muscle cells of ischemic myocardium

In subsequent studies, we evaluated the effects of chronic coronary occlusion and exercise training on whole-cell K⁺ currents in freshly isolated coronary arteriolar smooth muscle cells from nonoccluded and collateral-dependent myocardial regions (Xie, Parker & Heaps, 2013). We revealed that K⁺ channel currents were significantly decreased in arteriolar smooth muscle cells from collateral-dependent compared with nonoccluded regions in both sedentary and exercise-trained pigs. Interestingly, after exercise training, cells from arterioles in the collateral-dependent region displayed significantly reduced whole cell K⁺ channel currents compared with those in sedentary pigs and a strong tendency (p = 0.07) for reduced K⁺ channel currents in the nonoccluded region also was apparent after exercise. Thus, both chronic occlusion and exercise training resulted in a reduction in whole cell K⁺ channel currents, which was further diminished in the presence of occlusion plus exercise training. In contrast with these findings, smooth muscle cells isolated from conduit arteries demonstrated that neither exercise training nor chronic coronary occlusion altered whole cell K⁺ currents. Previous studies have shown unique exercise training-induced adaptations of K⁺ channel currents in coronary arteries compared with coronary arterioles of diabetic, dyslipidemic pigs (Mokelke et al., 2005; Mokelke et al., 2003). These studies have reported a reversal of increased K⁺ channel current in conduit smooth muscle cells after exercise training (Mokelke et al., 2003) and an increase in spontaneous transient outward currents in smooth muscle of intact arteriole segments after exercise (Mokelke et al., 2005). Similar to our findings in control and collateral-dependent conduit arteries (Deer & Heaps, 2013), other investigators using healthy control animals, have shown no effect of exercise training on whole-cell K⁺ channel currents in smooth muscle cells isolated from conduit coronary arteries, although both BKCa and 4-AP-sensitive Kv channels contributed more to basal tone in exercise-trained compared with sedentary animals (Bowles, Laughlin, et al., 1998).

Vascular smooth muscle cells in the coronary circulation express a wide range of Kv channels, with the Kv1, Kv2, and Kv7 families regarded as the most significant (Jackson, 2018). The wide variety of Kv channel isoforms likely makes them suitable end effectors in response to many different stimuli. Variations in targets across the different Kv channel isoforms may be due to differential subcellular localization or different binding sites located on specific channels. It has been recently proposed that the Kv7 family is involved in ischemia-induced vasodilation by means of H₂S and adenosine stimulation in a porcine model (Hedegaard et al., 2014). However, others have reported that Kv7 channels are not involved in metabolic or ischemic regulation of coronary vasodilation, but instead contribute to paracrine regulation via endothelium dependent mechanisms in swine (Goodwill et al., 2016). Interestingly, exercise training reversed a reduction in Kv1.2 and Kv1.5 channel expression in spontaneously hypertensive rats (Li, Lu, & Shi, 2014). Nishijima previously reported a loss of 4AP-sensitive Kv channel contribution to dilation in adipose tissue arterioles from coronary artery disease patients compared with those from non-diseased patients (Nishijima et al., 2017). The observed reduction in Kv channel contribution to dilation was attributed to a loss of Kv1.5 isoform in this study and a shift to BKCa as the predominant end effector in diseased patients (Nishijima et al., 2017). The same group also reported a reduced role for 4AP-sensitive Kv channels to dilation, and specifically the Kv1.5 isoform, in coronary arterioles isolated from diseased patients compared with those from patients not suffering from coronary artery disease (Nishijima et al., 2018).

5. BKCa channels in arteriolar smooth muscle cells of ischemic myocardium

It has been repeatedly shown that BKCa channels are abundantly expressed in vascular smooth muscle cell membranes and that they act as a negative feedback to increased intracellular calcium concentrations to reduce arterial tone (Berkefeld et al., 2010; Hill et al., 2010; Rusch, 2009; Toro et al., 2014). Under physiological conditions, receptor-mediated vasoconstriction is reduced through increases in BKCa channel activity (Albarwani, Al-Siyabi, Baomar, & Hassan, 2010; Bowles, Laughlin, et al., 1998; Laughlin, Bowles, & Duncker, 2012). Closely related to our work, previous studies in coronary arterioles of coronary artery disease patients demonstrated that both flow-mediated (Miura et al., 2001) and H₂O₂-induced (Liu, Bubolz, Mendoza, Zhang, & Gutterman, 2011; Zhang et al., 2012) dilation are mediated in part by BKCa channels. We were interested in exploring if the contribution of BKCa channels to endothelium-dependent dilation was an adaptation of the diseased state since this signaling pathway was not explored previously in non-diseased arterioles and whether the exercise training-enhanced dilation observed previously by our laboratory (Xie et al., 2013) was attributable to BKCa channel activation.

We examined a role for BKCa channels in endothelium-dependent dilation of the coronary microvasculature as well as that of large conduit coronary arteries. In cannulated coronary arterioles, endothelium-dependent dilation was impaired in the collateral-dependent region of ameroid-occluded swine (Xie, et al., 2013). Exercise training restored the impaired dilation, which we revealed was at least partially dependent on a greater contribution of BKCa channels after exercise. Previous literature also shows that increases in BKCa channel activity is brought about by exercise training in a porcine model of heart failure (Olver et al., 2018). We also evaluated whole-cell iberiotoxin-sensitive K⁺ channel current in freshly isolated coronary arteriolar smooth muscle cells from nonoccluded and collateral-dependent myocardial regions. These data revealed no differences in iberiotoxin-sensitive currents. Using immunoblot analysis, we have also discovered that protein levels for BKCa channel α-subunit did not differ in arterioles from nonoccluded and collateral-dependent arterioles of sedentary and exercise-trained swine (data not published). The lack of differences in BKCa channel currents and BKCa channel protein levels across treatment groups suggest that the increased contribution of BKCa channels after exercise in collateral-dependent arterioles may reflect better coupling of BKCa channels with second-messenger signaling pathways (e.g., Ca²⁺) that utilize BKCa channel as an end effector. We presume similar adaptations with Kv channel subtypes, which will be an area of future study in our laboratory.

We also demonstrated in freshly isolated smooth muscle cells of conduit coronary arteries there were no differences in iberiotoxin-sensitive K⁺ currents (Deer & Heaps, 2013). Interestingly, we demonstrate for the first time in Fig. 5 the comparison of K⁺ channel currents in smooth muscle cells isolated from conduits and arterioles. These comparisons reveal that total whole-cell K⁺ channel currents in smooth muscle cells isolated from conduit arteries are relatively smaller than those from arterioles in both nonoccluded (Fig. 5A) and collateral-dependent (Fig. 5B) vessels. In contrast, iberiotoxin-sensitive K⁺ currents are relatively greater in conduit than arteriole smooth muscle cells of both nonoccluded (Fig. 5C) and collateral-dependent (Fig. 5D) vessels. Taken together, these data suggest that Kv channels may play a more prominent role in coronary arterioles while BKCa channels may play a more prominent role in larger arteries. It should be noted that in the experiments performed for Fig. 5, pipettes were filled with a solution containing high EGTA (10mM), known to chelate intracellular Ca²⁺ and thereby minimize BKCa channel activity to that stimulated by changes in voltage; thus, while currents are comparable, the iberiotoxin-sensitive K⁺ currents are relatively low.

Fig. 5.

Effect of chronic occlusion and exercise training on A and B: whole-cell K⁺ channel current and C and D: iberiotoxin-sensitive K⁺ channel current in smooth muscle cells from coronary arteries and arterioles. Currents were elicited by 500-ms step depolarizations to test potentials (TP) ranging from −70 to +100 in 10-mV increments from a holding potential of −80 mV. Current-voltage relationships reveal relatively smaller whole-cell K⁺ channel currents in smooth muscle cells isolated from conduit compared with arterioles in nonoccluded (A) and collateral-dependent (B) vessels. In contrast, iberiotoxin-sensitive K⁺ currents are relatively larger in conduit than arteriole smooth muscle cells of both nonoccluded (C) and collateral-dependent (D) vessels. Current-voltage relationships were obtained by plotting the mean outward current at the end of each test potential normalized to cell membrane capacitance. Values are means ± SEM; n, number of animals, followed by number of cells. The legend key on left represents panels A and C;legend key on right represents panel B and D. Data were adapted from Deer, R. R., & Heaps, C. L. (2013). Exercise training enhances multiple mechanisms of relaxation in coronary arteries from ischemic hearts. American Journal of Physiology. Heart and Circulatory Physiology, 305, H1321-H1331.; W. Xie, W., Parker, J. L., & Heaps, C. L. (2013). Exercise training-enhanced, endothelium-dependent dilation mediated by altered regulation of BKCa channels in collateral-dependent porcine coronary arterioles. Microcirculation, 20(2), 170–182.

Interestingly, despite no exercise training-induced alterations in BKCa channel currents, these channels did contribute significantly to exercise training-enhanced, endothelium-dependent dilation of the coronary microvasculature (Xie et al., 2013). In cannulated coronary arterioles, endothelium-dependent, bradykinin-mediated dilation was impaired in the collateral-dependent region of ameroid-occluded swine (Xie et al., 2013). Exercise training restored the impaired dilation, which we revealed was at least partially dependent on a greater contribution of BKCa channels after exercise (Xie et al., 2013). We also demonstrated in coronary arteries, that although bradykinin-mediated relaxation was not altered by occlusion or exercise training under control conditions, relaxation was more persistent after inhibition of the endothelium-derived signaling molecules, nitric oxide and prostacyclin, after exercise training (Deer & Heaps, 2013). Inclusion of the BKCa channel blocker, iberiotoxin, with inhibitors of nitric oxide synthase and prostanoids abolished the enhanced exercise-training-mediated relaxation in nonoccluded and collateral-dependent arteries, such that relaxation responses were similar across treatments (Deer & Heaps, 2013). These findings suggested redundancy in signaling pathways of vascular relaxation after exercise training and that BKCa channels act in a compensatory role when other signaling molecules are impaired. In addition, our laboratory as well as others have reported involvement of BKCa channels in coronary vascular recovery and enhanced relaxation after exercise training in cardiovascular disease models (Deer & Heaps, 2013; Xie et al., 2013).

Recent data from our laboratory have revealed that exercise training restores impaired endothelium-dependent dilation in collateral-dependent arterioles through cellular adaptations that increase endothelium-derived H₂O₂ bioavailability (Xie, Parker, & Heaps, 2012). Our subsequent data indicate that BKCa and 4-aminopyridine-sensitive Kv channels are downstream effectors that contributed to enhanced H₂O₂-mediated dilation in the collateral-dependent arterioles after exercise training ( Johnson, Bray, & Heaps, 2021). Using confocal microscopy, we also demonstrated that H₂O₂ treatment increased colocalization of protein kinase A and BKCa channel proteins at the plasma membrane of smooth muscle cells isolated from collateral-dependent arterioles of exercise trained pigs (Johnson et al., 2021). These findings implicate increased PKA activation of BKCa channels as the signaling pathway that mediates the enhanced H₂O₂-induced dilation after exercise training in arterioles isolated from ischemic myocardium. We postulate that PKA activation of Kv channels may also play a role in the enhanced dilation observed after exercise training. Future studies will explore the contribution of specific Kv channel isoforms as downstream effectors of enhanced H₂O₂-mediated dilation after exercise training.

6. Role of additional K⁺ channel subfamilies in control of basal tone and vasodilation

Our laboratory has not explored the effect of chronic occlusion and exercise training on KATP channels in the coronary microcirculation. It has been shown that inhibition of KATP channels leads to increased basal tone, suggesting a significant contribution of these channels to the regulation of resting coronary blood flow (Duncker, Van Zon, Altman, Pavek, & Bache, 1993; Duncker, van Zon, Pavek, Herrlinger, & Bache, 1995; Merkus, Sorop, Houweling, Hoogteijling, & Duncker, 2006; Samaha, Heineman, Ince, Fleming, & Balaban, 1992). In human patients, Farouque, Worthley, Meredith, Skyrme-Jones, and Zhang (2002) assessed coronary hemodynamics before and after KATP blockade with intracoronary delivery of glibenclamide. In this study, glibenclamide significantly reduced resting coronary blood flow by 17%, but did not alter coronary flow reserve. However, while KATP channels play a significant role in determination of basal tone in swine and humans, several studies have demonstrated that they are not necessary for vasodilation associated with exercise hyperemia in animal models (Duncker, Oei, Hu, Stubenitsky, & Verdouw, 2001; Duncker et al., 1993; Duncker et al., 1995; Merkus et al., 2003) or adenosine-induced coronary hyperemia in humans (Farouque et al., 2002). We previously reported no role for KATP channels in adenosine-mediated dilation of coronary arterioles isolated from control male or female swine (Heaps & Bowles, 2002). Other investigators suggest a role for KATP channels after transient ischemia, i.e., reactive hyperemia (Berwick et al., 2010) and under hypoxic conditions (Liu & Flavahan, 1997), but the role of these channels in animal models or human patients with critical coronary stenosis is unclear.

In contrast to BKCa channels, small-conductance (SKCa) and intermediate-conductance (IKCa) are predominantly expressed within endothelial cells and have been shown to demonstrate higher sensitivity to calcium in comparison to BKCa channels (Burnham et al., 2002). Substantial literature support endothelial IKCa and SKCa channels as contributors to endothelium-derived hyperpolarization and vascular tone regulation (Levi-Rosenzvig et al., 2017; Liu et al., 2015). Additional studies have also revealed the expression of IKCa channels in vascular smooth muscle, however, this occurs primarily within proliferating smooth muscle cells (Bi et al., 2013; Tharp, Wamhoff, Turk, & Bowles, 2006; Zhao, Su, Wang, Li, & Deng, 2013), playing a role in the pathogenesis of cardiovascular disease and restenosis. Yet, Gutterman and colleagues (Sato et al., 2005) reported a significant role of smooth muscle IKCa channels in adenosine-mediated dilation with no significant contribution of BKCa, SKCa, KATP, or 4-AP-sensitive Kv channels in coronary arterioles of diseased patients. Within endothelial cells, Tune and colleagues (Kurian, Berwick, & Tune, 2011) found that activation of coronary microvascular IKCa channels resulted in increased coronary blood flow via nitric oxide production, whereas inhibition of the IKCa channels resulted in impaired coronary flow regulation. However, inhibition of these channels did not contribute to dilation following brief occlusion. In a hypoxic environment, other investigators found reduced endothelial cell K⁺ currents, mainly involving reduced SKCa and IKCa channels in porcine coronary arteries (Yang, Huang, Man, Yao, & He, 2011). These reductions were accompanied by decreased IKCa channel protein levels. Thus, SKCa and IKCa channels contribute mostly to endothelium-derived vasodilation with a negative impact of hypoxia; adaptations to chronic ischemia have not been explored.

7. Conclusion

This review has primarily focused on adaptations in K⁺ channel activity and contributions to vasodilation in the microcirculation distal to chronic coronary artery occlusion. We propose that K⁺ channel activity is a highly significant player in the dynamic regulation of coronary blood flow under physiological conditions as well as in adaptations to disease and exercise training. Despite our focus on adaptations in cellular signaling in the microcirculation distal to stenosis, we postulate that some of the observed alterations may be applicable to coronary microvascular dysfunction in the absence of obstructive coronary artery disease. Further research and insight into mechanisms that control coronary blood flow in health and disease, as well as exercise training will support novel and potentially more effective therapeutic approaches for the prevention and treatment of cardiovascular disorders.