Adenosine Prevents Hyperkalemia-Induced Calcium Loading in Cardiac Cells: Relevance for Cardioplegia

Transcription

1 Adenosine Prevents Hyperkalemia-Induced Calcium Loading in Cardiac Cells: Relevance for Cardioplegia Aleksander Jovanović, MD, Alexey E. Alekseev, PhD, JóseR.López, MD, Win K. Shen, MD, and Andre Terzic, MD Division of Cardiovascular Diseases and Departments of Internal Medicine and Pharmacology, Mayo Clinic, Mayo Foundation, Rochester, Minnesota Background. Hyperkalemic cardioplegic solutions effectively arrest the heart but also induce membrane depolarization, which could lead to intracellular Ca 2 loading and contribute to ventricular dysfunction associated with cardiac operations. Adenosine, which possesses cardioprotective properties, has been proposed as an adjunct to conventional cardioplegic solutions. However, it is not known whether adenosine supplementation enables cardiac cells to withstand hyperkalemiainduced Ca 2 loading. Methods. Single ventricular cardiomyocytes were isolated from guinea pig hearts, loaded with a Ca 2 - sensitive fluorescent probe, and imaged by digital epifluorescent microscopy. The emitted fluorescence of the probe, a measure of the intracellular Ca 2 concentration, was recorded from single myocytes during hyperkalemic challenges in the absence and the presence of adenosine to assess the protective effectiveness of this agent. Results. Hyperkalemic solutions induced intracellular Ca 2 loading (estimated intracellular Ca 2 concentration, 88 5 nmol/l before and 1, nmol/l after addition of 16 mmol/l KCl). Adenosine (1 mmol/l) prevented K -induced Ca 2 loading (intracellular Ca 2 concentration, 86 6 nmol/l before and 85 8 nmol/l after exposure to K ). Whereas glyburide (3 mol/l), an antagonist of adenosine triphosphate sensitive K channels, had no effect, staurosporine (200 nmol/l) and chelerythrine (5 mol/l), two inhibitors of protein kinase C, did abolish the action of adenosine. Conclusions. Adenosine prevents hyperkalemia-induced Ca 2 loading in cardiomyocytes. This effect is due to a direct action on ventricular cells, as the preparation employed was free from atrial, neuronal, and vascular elements, and appears to be mediated through a protein kinase C dependent mechanism. The property of adenosine to prevent hyperkalemia-induced Ca 2 loading may contribute to the cytoprotective efficacy of this agent as an adjunct to conventional hyperkalemic cardioplegic solutions. (Ann Thorac Surg 1997;63:153 61) 1997 by The Society of Thoracic Surgeons Accepted for publication July 24, Address reprint requests to Dr Terzic, Guggenheim 7F, Mayo Clinic, Rochester, MN ( terzic.andre@mayo.edu). Hyperkalemic cardioplegic solutions effectively arrest the heart during open heart operations by depolarizing the sarcolemma. However, a recognized adverse effect of hyperkalemic cardioplegia is the possible development of ventricular dysfunction believed to be related, in part, to intracellular Ca 2 loading, a consequence of K -induced membrane depolarization [1 4]. In recent years, adenosine has been shown to possess cardioprotective properties under conditions in which the myocardium is exposed to elevated concentrations of K, such as cardioplegic arrest and global surgical ischemia [5 8]. It has been recognized in clinical practice that adenosine as a supplement to conventional hyperkalemic cardioplegic solutions is beneficial in reducing the risk of contractile dysfunction after cardiac surgical procedures [7, 9, 10]. However, it remains unknown whether adenosine can protect the myocardium against K -induced Ca 2 loading. Previously described adenosine-mediated cardioprotective mechanisms include an antiadrenergic action, a reduction in the degradation of adenosine triphosphate (ATP) during global ischemia, an improvement in ATP repletion during reperfusion, an inhibition of the adherence of stimulated neutrophils to endothelial cells, an inhibition of platelet aggregation, an inhibition of neutrophil-induced generation of superoxide anion and hydrogen peroxide, a decrease in oxygen demand, and an increase in oxygen supply because of coronary vasodilatation [5, 8, 11]. Overtly, none of these mechanisms can account for the possible direct protective action of adenosine on cardiac myocytes that might reduce K -induced intracellular Ca 2 loading. Recently, it has been shown that opening of ATP sensitive K channels, translocation of protein kinase C, or both could prevent Ca 2 loading and could play a protective role in cardiac myocytes exposed to elevated K concentrations or during ischemic preconditioning [3, 1997 by The Society of Thoracic Surgeons /97/$17.00 Published by Elsevier Science Inc PII S (96)

2 154 JOVANOVIĆ ET AL Ann Thorac Surg ADENOSINE AND K -INDUCED Ca 2 LOADING 1997;63: , 12 16]. Under certain conditions, adenosine can activate both ATP sensitive K channels and protein kinase C in cardiac cells [17, 18]. Therefore, the aims of the present study were to examine whether adenosine protects single ventricular cells against Ca 2 loading induced by high extracellular K concentration and to assess the contribution of ATP sensitive K channels, protein kinase C activation, or both to the possible cytoprotective action of adenosine when used as a supplement to hyperkalemic solutions. Material and Methods Isolation of Single Ventricular Cardiomyocytes Ventricular myocytes were isolated by enzymatic dissociation [17]. In brief, guinea pigs were anesthetized with sodium pentobarbital and artificially ventilated. The aorta was cannulated in situ, and after cardiotomy, the heart was retrogradely perfused (at 37 C) with the following solutions: first, control bathing solution (in millimoles per liter: NaCl, 136.5; KCl, 5.4; CaCl 2, 1.8; MgCl 2, 0.53; glucose, 5.5; HEPES-NaOH, 5.5; ph 7.4) for 5 to 10 minutes; second, nomimally Ca 2 -free bathing solution; third, nominally Ca 2 -free bathing solution containing collagenase (0.04 g/100 ml, Sigma type I) for 30 minutes; and fourth, high-k, low-cl solution (in millimoles per liter: taurine, 10; oxalic acid, 10; glutamic acid, 70; KCl, 25; KH 2 PO 4, 10; glucose, 11; EGTA, 0.5; HEPES-KOH, 10; ph 7.4) for 5 minutes. Ventricles were separated from atria and stored in the high-k, low-cl solution at 4 C. Single cells were isolated by agitating a small piece of dissected ventricle in a culture dish filled with the control bathing solution. The investigation conformed with the Guide for the Care and Use of Laboratory Animals (NIH publication 85-23, revised 1985), and the experimental protocol was approved by the Institutional Animal Care and Use Committee at the Mayo Clinic. Digital Epifluorescent Microscopy Ventricular myocytes were loaded with 3.5 mol/l of the Ca 2 -selective fluorescent probe, Fluo-3 acetoxylmethyl ester (Fluo-3AM), for 20 minutes at room temperature (21 to 23 C) as previously described [3, 19]. This probe exhibits a lower binding capacity for Ca 2 and produces a larger fluorescence signal after Ca 2 binding than conventional fluorescent probes. Myocytes loaded with Fluo-3AM were transferred to an experimental chamber mounted on the stage of the epifluorescent microscope and permitted to adhere to the glass bottom of the chamber. In a separate set of experiments, cardiomyocytes were loaded with the ratiometric dye Fura-2 acetoxylmethyl ester (Fura-2AM) (10 mol/l) to quantify resting intracellular Ca 2 concentration. Ventricular myocytes were imaged at 21 to 23 C by digital epifluorescent microscopy using an inverted microscope (Zeiss Axiovert-135 TV) with a 40 oilimmersion objective lens. Optimal focus was adjusted by viewing the myocytes under bright field microscopy. A 100 W mercury lamp served as a source of light to excite Fluo-3AM at 488 nm or Fura-2AM at 340 and 380 nm. After crossing a dichroic mirror, fluorescence emitted at 520 nm by the excited dyes was captured by an intensified-charge coupled-device camera and digitized using the epifluorescent imaging system (Attoflor Ratio Vision). Background fluorescence (Tyrode s solution containing no cells) was subtracted from the fluorescence of Fluo- 3AM or Fura 2AM loaded myocytes. Calibration of Fura-2AM and Fluo-3AM Signals In cells loaded with Fura-2AM, an estimate of the Ca 2 concentration ([Ca 2 ]) was obtained according to the equation Ca 2 R R min /R max R K d where R is the fluorescence ratio recorded from the cell; R min and R max represent the fluorescence ratio in the absence of Ca 2 (extracellular Ca 2 removed and 3 mmol/l EGTA added to the extracellular solution) and at high [Ca 2 ] (3 mmol/l CaCl 2 ), respectively; K d is the Ca 2 dissociation constant of the dye (236 nmol/l); and is the ratio of R min /R max at 380 nm. To obtain R min and R max, Fura-2AM loaded cardiac cells were exposed to the calcium ionophore 4-bromo A To prevent cell contraction in permealized cells exposed to high concentrations of extracellular Ca 2, myocytes were pretreated with carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (2 mol/l) and 2,3-butaneodione monoxime (40 mmol/l). An estimate of the increase in [Ca 2 ] i as a function of Fluo-3AM fluorescence was calculated by resolving the system of three equations that include the expression for equilibrium Ca 2 concentration using the relative fraction of bound (f Ca ) and unbound (f u ) Fluo-3AM to Ca 2 : Ca 2 K d f Ca /f u (1) where K d is the dissociation constant of Fluo-3AM (526 nmol/l), and f Ca and f u relate to the total concentration of Fluo-3AM by the relation f Ca f u f t (2) The relation between Fluo-3AM intensity (F) and the variables just listed is given by F F max f Ca /f t F min f u /f t (3) where F max and F min are the maximum and minimum Fluo-3AM fluorescence intensity. Resolving equations 1, 2, and 3 relative to [Ca 2 ] produced the following equation: Ca 2 K d F F min /F max F The estimate of cytosolic Ca 2 concentration was calculated taking into account the resting cytosolic Ca 2 concentration: Ca 2 K d F F min /F max F Ca 2 r where [Ca 2 ] r is the value of resting cytosolic Ca 2 concentration estimated from Fura-2AM experiments (see above) and F min F F max.

3 Ann Thorac Surg JOVANOVIĆ ET AL 1997;63: ADENOSINE AND K -INDUCED Ca 2 LOADING 155 Fig 1. Elevation of extracellular K induces intracellular Ca 2 loading in single ventricular cardiomyocytes. (A) Epifluorescent images from a Fluo-3 acetoxylmethyl ester (Fluo- 3AM) loaded cardiomyocyte prior to (frame 1) and after the (frame 2) addition of 16 mmol/l K to the solution bathing the myocyte. White horizontal bar indicates 20 m. (A1) Changes in intensity of Fluo- 3AM fluorescence plotted as a function of time from myocyte in Figure 1A. Circles labeled 1 and 2 correspond to frames 1 and 2. (B) Average changes in maximal fluorescence under control conditions, and after addition of 16 mmol/l K. Each bar represents the mean the standard error of the mean (* p 0.01 between the two means.) Experimental Protocol To examine the effect of K on cytosolic Ca 2 concentration, myocytes were exposed first to Tyrode s solution (control) and then to that solution with 16 mmol/l KCl added (so that the total K concentration in the solution bathing a myocyte was 21.4 mmol/l). To assess the effect of adenosine, myocytes were exposed first to Tyrode s solution (control); then adenosine was added (100 mol/l or 1 mmol/l), followed by adenosine (100 mol/l or 1 mmol/l) plus 16 mmol/l KCl, and finally, only 16 mmol/l KCl. Unless otherwise indicated, to determine the effect of isobutyl-methylxanthine, glyburide, staurosporine, or chelerythrine, myocytes were first incubated for 10 to 30 minutes in Tyrode s solution supplemented with isobutyl-methylxanthine (100 mol/l), glyburide (3 mol/l), staurosporine (200 nmol/l), or chelerythrine (5 mol/l), which served as respective controls. Then in the continuous presence of glyburide (3 mol/l), staurosporine (200 nmol/l), or chelerythrine (5 mol/l), adenosine was added (1 mmol/l), followed by adenosine (1 mmol/l) plus 16 mmol/l K, and finally, only 16 mmol/l K. These experimental protocols were selected so that each cell served as its own control. Data obtained with cells that did not respond by intracellular Ca 2 loading to the final exposure to 16 mmol/l K were excluded from the analysis. Drugs Adenosine, glyburide, staurosporine, and chelerythrine were from Sigma Chemical Co, whereas Fluo-3AM and Fura-2AM were purchased from Molecular Probes. Adenosine and chelerythrine were dissolved in Tyrode s solution and diluted prior to each experiment. Glyburide, staurosporine, and Fura-2AM were dissolved in dimethyl sulfoxide, and Fluo-3AM was dissolved in dimethyl sulfoxide plus pluronic acid. All substances were diluted in Tyrode s solution prior to each experiment. The final concentration of dimethyl sulfoxide in Tyrode s solution was less than 0.1%. Statistical Analysis Results are expressed as the mean the standard error of the mean; n refers to the number of experiments. Significant differences between two means were determined with the Student t test for paired or unpaired observations where appropriate. A p value of less than 0.05 was considered significant. Results Effect of Hyperkalemic Challenge on Cytosolic Ca 2 Concentration in Single Ventricular Cells At rest, ventricular cardiomyocytes had an estimated cytosolic Ca 2 concentration of 88 5 nmol/l (n 52)

4 156 JOVANOVIĆ ET AL Ann Thorac Surg ADENOSINE AND K -INDUCED Ca 2 LOADING 1997;63: Fig 2. Adenosine prevents K -induced Ca 2 loading. (A) Epifluorescent images from a cardiomyocyte under control conditions (frame 1), in adenosine (frame 2), in adenosine plus 16 mmol/l K (frame 3), and in 16 mmol/l K only (frame 4). White horizontal bar indicates 20 m. (A 1 ) Changes in intensity of Fluo-3 acetoxymethyl ester (Fluo-3AM) fluorescence plotted as a function of time from myocyte in Figure 2A. Circles labeled 1 through 4 correspond to frames 1 through 4. Adenosine alone did not significantly change the Fluo-3AM fluorescence, yet it prevented 16 mmol/l K to induce Ca 2 loading. (B) Average changes in maximal fluorescence under control conditions, in adenosine, in adenosine plus 16 mmol/l K, and in 16 mmol/l K only. Each bar represents the mean the standard error of the mean. (* p 0.01 compared with control.) (Fig 1A, frame 1, 1B). Exposure of cardiomyocytes to a K -rich (plus 16 mmol/l KCl; final concentration, 21.4 mmol/l) solution increased cytosolic Ca 2 concentration to 1, nmol/l (n 52) (see Fig 1A, frame 2, 1B). This increase was significant compared with the values of cytosolic Ca 2 concentration obtained before exposure to this elevated concentration of extracellular K (p 0.01) (Fig 1). Under our experimental conditions, the peak level of intracellular Ca 2 loading was induced within 57 4 seconds (n 52) after bathing the cardiomyocytes with the K -rich solution. Effect of Adenosine on K -Induced Ca 2 Loading Exposure of resting single ventricular cells to adenosine (1 mmol/l) did not affect cytosolic Ca 2 concentration (Figs 2A, 2B). The estimated Ca 2 concentration in cardiomyocytes was 89 9 nmol/l in the absence and 86 6 nmol/l in the presence of adenosine (p 0.05) (n 27) (see Fig 2B). However, in the continuous presence of adenosine (1 mmol/l), exposure of cardiomyocytes to K -rich solution did not induce an increase in cytosolic Ca 2 (see Fig 2A, 2B). The estimated concentration of Ca 2 in the presence

5 Ann Thorac Surg JOVANOVIĆ ET AL 1997;63: ADENOSINE AND K -INDUCED Ca 2 LOADING 157 of 1 mmol/l adenosine plus 16 mmol/l K was 85 8 nmol/l (n 27; see Fig 2B). This value was not significantly different from the value obtained before application of the elevated K concentration either in the absence (n 27) or presence of adenosine (n 27). In ventricular cells continuously exposed to the elevated K concentration, washout of adenosine resulted in a significant increase (p 0.01) in the cytosolic Ca 2 concentration to 1, nmol/l (n 27) (see Figs 2A, 2B). Exposure of cardiomyocytes to a lower concentration of adenosine (100 mol/l) did not prevent intracellular Ca 2 loading induced by the elevated K concentration, as the estimated concentrations of Ca 2 in the presence of adenosine (100 mol/l) and adenosine (100 mol/l) plus 16 mmol/l K were 96 6 nmol/l and 2, nmol/l, respectively (n 9) (p 0.01). Thus, it is at a concentration of 1 mmol/l that adenosine is effective in preventing intracellular Ca 2 loading induced by the high extracellular K concentration. Effect of Isobutyl-Methylxanthine on Action of Adenosine on K -Induced Ca 2 Loading Cardiomyocytes were treated with isobutyl-methylxanthine (100 mol/l), a nonselective inhibitor of adenosine receptors. In all myocytes so tested, isobutyl-methylxanthine did not significantly affect the ability of adenosine (1 mmol/l) to prevent extracellular K -induced Ca 2 loading, as the estimated Ca 2 concentration was 96 6 nmol/l before and nmol/l after 16 mmol/l K was added to cardiomyocytes pretreated with and continuously exposed to 100 mol/l isobutylmethylxanthine plus 1 mmol/l adenosine (n 4) (not illustrated). Thus, the action of adenosine appears to be mediated through an isobutyl-methylxanthine insensitive mechanism. Effect of Glyburide, Staurosporine, and Chelerythrine on Action of Adenosine In several experimental models, it has been shown both in vivo and in vitro that the opening of ATP sensitive K channels contributes to the cardioprotective action of low concentrations of adenosine. Under our experimental conditions, glyburide (3 mol/l), an antagonist of these K channels, did not affect the ability of adenosine (1 mmol/l) to protect single ventricular cells against K - induced Ca 2 loading (Fig 3A). In glyburide-pretreated myocytes, the estimated cytosolic Ca 2 concentration was 86 9 nmol/l in the presence of adenosine versus nmol/l (n 6) in the presence of adenosine plus 16 mmol/l K (p 0.05) (Fig 3C; see Fig 3A). Recently, it has been demonstrated that the cardioprotective action of adenosine may be mediated through a staurosporine-sensitive mechanism. In contrast to glyburide, staurosporine (200 nmol/l) completely abolished the cytoprotective action of 1 mmol/l adenosine (Fig 3B). In staurosporine-treated myocytes, the estimated value of the cytosolic Ca 2 concentration was 87 6 nmol/l in the presence of adenosine and 1, nmol/l in the presence of adenosine plus 16 mmol/l K (see Figs 3B, 3C). The difference between these values was significant (p 0.01). Thus, the protective effect of adenosine on K -induced Ca 2 loading appears to be mediated through a staurosporine-sensitive mechanism. As staurosporine is a nonselective kinase inhibitor, we tested the effect of chelerythrine, a rather selective inhibitor of protein kinase C, on the action of adenosine. Similar to staurosporine, pretreatment and continuous exposure of cardiomyocytes to chelerythrine (5 mol/l) blocked the effect of adenosine (1 mmol/l) on extracellular K -induced Ca 2 loading (Fig 4). In the cardiomyocyte shown in Figure 4A, adenosine (1 mmol/l) initially prevented 16 mmol/l K from elevating intracellular Ca 2 concentration (see Fig 4A, frames 1 3, Fig 4A 1 ). However, treatment of the same myocyte with chelerythrine (5 mol/l) for 10 minutes abolished the protective effect of adenosine on K -induced Ca 2 loading (see Fig 4A, frames 4 6, Fig 4A 1 ). On average, in 5 mol/l chelerythrine treated cardiomyocytes, the estimated Ca 2 concentration was 95 8 nmol/l and 1, nmol/l in adenosine (1 mmol/l) and adenosine (1 mmol/l) plus 16 mmol/l K, respectively (n 8) (p 0.01) (see Fig 4B). Thus, chelerythrine inhibits adenosinemediated protection against extracellular K induced Ca 2 loading in cardiomyocytes. Comment K -Induced Ca 2 Loading In the present study, we exposed cardiomyocytes to an elevated extracellular K concentration, a challenge associated with cardioplegic cardiac arrest. We observed that extracellular K at higher than 16 mmol/l, a concentration to which myocardial cells are commonly exposed during cardioplegia [9], induced significant intracellular Ca 2 loading. The direct visualization of the effect of K at the single cell level is in accord with previous studies performed in suspensions of cardiac cells, which have also shown that elevation of extracellular K induces Ca 2 loading [1, 2]. This is an important finding in view of the notion that intracellular Ca 2 loading of cardiomyocytes has been considered to represent a precipitating factor that could lead to diastolic dysfunction and cell impairment [4]. Adenosine Prevents K -Induced Ca 2 Loading We observed that adenosine effectively inhibited K - induced Ca 2 loading in ventricular myocytes. Although the property of adenosine to protect the myocardium against various insults has been recognized [5 10], one finding represents direct evidence at the single myocyte level that adenosine could interfere with intracellular Ca 2 loading. Thus, adenosine possesses a cytoprotective property in ventricular cardiomyocytes against K - induced Ca 2 loading that can occur in the absence of adrenergic costimulation. The protective effect of adenosine against K -induced Ca 2 loading is clearly related to its action on a myocardial cell, as the preparation employed was free from atrial, neuronal, and vascular elements.

6 158 JOVANOVIĆ ET AL Ann Thorac Surg ADENOSINE AND K -INDUCED Ca 2 LOADING 1997;63: Fig 3. Staurosporine, but not glyburide, antagonizes action of adenosine against K -induced Ca 2 loading. Epifluorescent images from (A) glyburide-pretreated cardiomyocyte and (B) staurosporine-pretreated cardiomyocyte under control conditions (frame 1), in adenosine (frame 2), in adenosine plus 16 mmol/l K (frame 3), and in 16 mmol/l K (frame 4) and corresponding graph showing changes in intensity of Fluo-3 acetoxylmethyl ester fluorescence as a function of time. Circles labeled 1 through 4 correspond to frames 1 through 4. Glyburide (3 mol/l) or staurosporine (200 nmol/l) was continuously present throughout the experiment. White horizontal bar indicates 20 m, and corresponding graph. (C) Average changes in maximal fluorescence under control conditions, in adenosine, in adenosine plus 16 mmol/l K, and in 16 mmol/l K in glyburide-pretreated (n 6) or in staurosporine-pretreated (n 4) cardiomyocytes. Each bar represents the mean the standard error of the mean. (* p 0.01) compared with respective controls.)

7 Ann Thorac Surg JOVANOVIĆ ET AL 1997;63: ADENOSINE AND K -INDUCED Ca 2 LOADING 159 Fig 4. Chelerythrine, an inhibitor of protein kinase C, antagonizes the action of adenosine against K -induced Ca 2 loading. (A) Epifluorescent images from adenosine-treated cardiomyocyte challenged with 16 mmol/l K before (frame 2) and after (frames 5 and 6) treatment with chelerythrine. White horizontal bar indicates 20 m. (A 1 ) Corresponding graph showing changes in intensity of Fluo-3 acetoxylmethyl ester fluorescence as a function of time. Circles labeled 1 through 6 correspond to frames 1 through 6. (B) Average changes in maximal fluorescence in chelerythrine-pretreated cardiomyocytes (n 8) in adenosine and in adenosine plus 16 mmol/l K. Each bar represents the mean the standard error of the mean. (arb. arbitrary; * p 0.01 compared with adenosine alone.) The cardioprotective potency of adenosine reported under various conditions is rather variable. Whereas micromolar concentrations are effective under ischemic conditions [5 8], millimolar concentrations are usually used when adenosine is an adjunct in potassium cardioplegia [9]. In the majority of studies, the cytoprotective action of adenosine has been associated with stimulation of adenosine A 1 receptors [5, 8]. In our study, the action of adenosine was insensitive to isobutyl-methylxanthine, a nonselective antagonist of adenosine receptors [20], which is in line with previous reports [21, 22] of a cardioprotective action of adenosine on myocardial functional recovery independent of receptor activation. Adenosine Triphosphate Sensitive K Channels Recently, it has been shown that openers of ATP sensitive K channels are protective [23, 24] under conditions associated with moderately elevated concentrations of extracellular K [3]. Although adenosine can activate ATP sensitive K channels in ventricular myocytes [17], the role of ATP sensitive K channels in adenosine-mediated cardioprotection remains controversial [23 25]. We applied sufficient concentrations of glyburide to block ATP sensitive K channels in cardiomyocytes, but glyburide did not modify the protective effect of adenosine on K -induced Ca 2 loading. This result does not support a relevant participation of ATP sensitive K channels in the cardioprotective effect of adenosine against K -induced Ca 2 loading, at least at the concentration of extracellular K used. It should be pointed out, however, that at concentrations of about 20 mmol/l extracellular K, the membrane potential closely approaches the value of the equilibrium potential for K, and it is hardly to be expected that an activator of ATP sensitive K channels will have major effects related to the opening of K channels [3]. Therefore, it appears that the possible participation of ATP sensitive K channels in myocardial protection is closely related to the nature of the challenge, including the concentration of extracellular K. Role for Protein Kinase C Under certain conditions, activation of protein kinase C has been related to a cardioprotective outcome [12 16]. Adenosine activates protein kinase C in heart cells [18], and inhibitors of protein kinase C can abolish the adenosine-evoked limitation of myocardial infarct size [26]. Staurosporine, a microbial alkaloid isolated from Streptomyces species, is a potent inhibitor of protein kinase C, with an inhibition constant in the nanomolar range [27]. In the present study, nanomolar concentrations of staurosporine effectively inhibited the protective effect of

8 160 JOVANOVIĆ ET AL Ann Thorac Surg ADENOSINE AND K -INDUCED Ca 2 LOADING 1997;63: adenosine on K -induced Ca 2 loading. Although staurosporine may also exhibit inhibitory efficacy on other types of protein kinases, the employed concentration of staurosporine is effective in inhibiting protein kinase C [27]. Moreover, the benzophenanthridine alkaloid chelerythrine, which selectively inhibits protein kinase C compared with tyrosine kinase, cyclic adenosine monophosphate dependent protein kinase, or calcium/ calmodulin-dependent kinase [28], also abolished the effect of adenosine against K -induced Ca 2 loading. Hence, the action of adenosine on K -induced Ca 2 loading in cardiomyocytes may be mediated through a staurosporine- and chelerythrine-sensitive mechanism, such as the activation of protein kinase C. Such a possibility is further supported by previous findings that activation of protein kinase C decreases cytosolic Ca 2 concentration in cardiomyocytes [29]. In this regard, it should be indicated that adenosine was found to slow the rate of K -induced membrane depolarization in single cardiomyocytes, which could translate into a net decrease in Ca 2 influx [30]. Adenosine, however, does not inhibit L-type Ca 2 channels per se [30], a finding suggesting that other mechanisms that participate in Ca 2 homeostasis and are regulated by protein kinase C [29] may contribute to the adenosine-induced protection of cardiomyocytes against K -induced Ca 2 loading. Further studies are required to define the intracellular signaling cascade that mediates the effect of adenosine in cardiomyocytes with an associated decrease in intracellular Ca 2 loading. Study Limitations To determine the direct action of adenosine on intracellular Ca 2 levels after a challenge with a high extracellular K concentration, it was necessary to image isolated cardiomyocytes. Although such an approach provided a direct visualization of the protective action of adenosine at the single cell level, it should be kept in mind that in the intact myocardium, additional cardiac and extracardiac mechanisms could modulate the cardioprotective efficacy of adenosine [11]. Also, it is conceivable that activation (or inhibition) of yet unrecognized intracellular signaling pathway may contribute to the action of adenosine described in this study. Conclusions The finding that adenosine prevents Ca 2 loading in ventricular myocardial cells exposed to elevated extracellular K concentrations may have important clinical implications for the adenosine-mediated protection of the ventricle. This concept deserves to be further tested in view of the reported protective efficacy of adenosine as a supplement to hyperkalemic cardioplegic solutions during open heart operations [7, 9]. A possible ramification of the present work is that agents known to act through protein kinase C may also have important cardioprotective properties under these conditions. This work was supported by the American Heart Association (Minnesota Affiliate), the Pharmaceutical Research and Manufacturers of America Foundation, and the Miami Heart Research Institute. References 1. Powell T, Tatham PER, Twist VW. Cytoplasmic free calcium measured by QUIN2 fluorescence in isolated ventricular myocytes at rest and during potassium-depolarization. Biochem Biophys Res Commun 1984;122: Cyran SE, Philips J, Ditty S, Baylen BG, Cheung J, LaNoue K. Developmental differences in cardiac myocyte calcium homeostasis after steady-state potassium depolarization: mechanisms and implications for cardioplegia. J Pediatr 1993;122:S López JR, Ghanbari R, Terzic A. An opener of ATP-sensitive K channels protects cardiomyocytes from moderate hyperkalemia-induced Ca 2 waves: a laser confocal microscopy study. Am J Physiol 1996;39:H Tani M. Mechanism of Ca 2 -overload in reperfused ischemic myocardium. Annu Rev Physiol 1990;52: Thornton JD, Liu GS, Olsson RA, Downey JM. Intravenous pretreatment with A1-selective adenosine analogues protects the heart against infarction. Circulation 1992;85: Toombs CF, McGee DS, Johnston WE, Vinten-Johansen J. Myocardial protective effects of adenosine. Infarct size reduction with pretreatment and continued receptor stimulation during ischemia. Circulation 1992;86: Vinten-Johansen J, Nakanishi K, Zhao ZQ, McGee DS, Tan P. Adenosine improves surgical myocardial protection with blood cardioplegia in ischemically injured canine heart. Circulation 1993;88(Suppl 2): Yao Z, Gross GJ. Glibenclamide antagonizes adenosine A1 receptor-mediated cardioprotection in stunned myocardium. Circulation 1993;88: De Jong JW, Der Meer PV, Loon HV, Owen P, Opie LH. Adenosine as adjunct to potassium cardioplegia: effect on function, energy metabolism, and electrophysiology. J Thorac Cardiovasc Surg 1990;100: Keller MW, Geddes L, Spotnitz W, Kaul S, Duling BR. Microcirculatory dysfunction following perfusion with hyperkalemic, hypothermic cardioplegic solutions and blood reperfusion. Effect of adenosine. Circulation 1991;84: Belardinelli L, Linden J, Berne RM. The cardiac effects of adenosine. Prog Cardiovasc Dis 1989;32: Armstrong S, Downey JM, Ganote CE. Preconditioning of isolated rabbit cardiomyocytes: induction by metabolic stress and blockade by the adenosine antagonist SPT and calphostin C, a protein kinase C inhibitor. Cardiovasc Res 1994;28: Liu Y, Ytrehus K, Downey JM. Evidence that translocation of protein kinase C is a key event during ischaemic preconditioning of rabbit myocardium. J Mol Cell Cardiol 1994;26: Baxter GF, Goma FM, Yellon DM. Involvement of protein kinase C in the delayed cytoprotection following sublethal ischaemia in rabbit myocardium. Br J Pharmacol 1995;115: Mitchell MB, Meng X, Ao L, Brown JM, Harken AH, Banerjee A. Preconditioning of isolated rat heart is mediated by protein kinase C. Circ Res 1995;76: Speechly-Dick ME, Mocanu MM, Yellon DM. Protein kinase C. Its role in ischemic preconditioning in the rat. Circ Res 1994;75: Terzic A, Tung RT, Inanobe A, Katada T, Kurachi Y. G proteins activate ATP-sensitive K channels by antagonizing ATP-dependent gating. Neuron 1994;12: Henry P, Demolombe S, Puceat M, Escande D. Adenosine stimulation activates delta protein kinase C in rat ventricular myocytes. Circ Res 1996;78: López JR, Jovanovic A, Terzic A. Spontaneous calcium waves without contraction in cardiac myocytes. Biochem Biophys Res Commun 1995;214:781 7.

9 Ann Thorac Surg JOVANOVIĆ ET AL 1997;63: ADENOSINE AND K -INDUCED Ca 2 LOADING Leung E, Johnston CI, Woodcock EA. Demonstration of specific receptors for adenosine in guinea-pig myocardium. Clin Exp Pharmacol Physiol 1983;10: Liu Y, Downey JM. Ischemic preconditioning protects against infarction in rat heart. Am J Physiol 1992;263:H Bolling SF, Childs KF, Ning XH. Adenosine s effect on myocardial functional recovery: substrate or signal? J Surg Res 1994;57: Grover GJ, Sleph PG, Dzwonczyk S. Role of myocardial ATP-sensitive potassium channels in mediating preconditioning in the dog heart and their possible interaction with adenosine A1-receptors. Circulation 1992;86: Yao Z, Gross GJ. The ATP-dependent potassium channel: an endogenous cardioprotective mechanism. J Cardiovasc Pharmacol 1994;24(Suppl 4):S Xu J, Wang L, Hurt CM, Pelleg A. Endogenous adenosine does not activate ATP-sensitive potassium channels in the hypoxic guinea pigs ventricle in vivo. Circulation 1994;89: Capogrossi MC, Kaku T, Filburn CR, et al. Phorbol ester and dioctanoylglycerol stimulate membrane association of protein kinase C and have a negative inotropic effect mediated by changes in cytosolic Ca 2 in adult rat cardiac myocytes. Circ Res 1990;66: Sakamoto J, Miura T, Goto M, Iimura O. Limitation of myocardial infarct size by adenosine A(1) receptor activation is abolished by protein kinase C inhibitors in the rabbit. Cardiovasc Res 1995;29: Budworth J, Gescher A. Differential inhibition of cytosolic and membrane-derived protein kinase C activity by staurosporine and other kinase inhibitors. FEBS Lett 1995;362: Herbert JM, Augereau JM, Gleye J, Maffrand JP. Chelerythrine is a potent and specific inhibitor of protein kinase C. Biochem Biophys Res Commun 1990;172: Alekseev AE, Jovanovic A, Lopéz JR, Terzic A. Adenosine slows the rate of K -induced membrane depolarization in ventricular cardiomyocytes: possible implication in hyperkalemic cardioplegia. J Mol Cell Cardiol 1996;28: The Annals of Thoracic Surgery Cumulative Index The Annals of Thoracic Surgery 31-year cumulative index, volume 1 through volume 60, January 1965 through December 1995, is now available in two versions: in print (ISBN ) and on CD-ROM (ISBN ). Both print and CD-ROM versions contain subject and author indexes for the 31 years of the journal to date. The CD-ROM also contains all of the journal s published scientific abstracts; hypertext links between article titles, subject headings, authors, and abstracts; a search function that allows full-text, Boolean, and keyword searches; and functions to select and format references for future use. The CD-ROM is both DOS/ Windows and Macintosh compatible. The price is $95.00 for the CD-ROM version, $95.00 for the print version, or only $ for both the CD-ROM and print versions. Contact Elsevier Science Inc to place your order: Telephone: (212) ; Fax: (212) by The Society of Thoracic Surgeons Ann Thorac Surg 1997;63: /97/$17.00 Published by Elsevier Science Inc PII S (96)