(:J "’""' ABSTRACT MECHANISM OF INOTROPIC ACTION OF CARDIAC GLYCOSIDES By David Dalee Ku It has been speculated for a long time that alterations of the transmembrane Na+ and K+ movements may be causally related to the positive inotropic action of digitalis. Recently, evidence has accu- mulated to suggest that the mechanism of the positive inotropic action of cardiac glycosides is related to the inhibition of cardiac Na+,K+- ATPase (MgZ+-dependent, (Na++K+)-activated ATP phosphohydrolase, EC 3.6.l.3), an enzyme system responsible for the active transport of Na+ and K+ across the cytoplasmic membrane. Nevertheless, a causal rela- tionship between these two phenomena is still not established because several investigators have reported that the inotropic responses of cardiac glycosides do not follow the time course of the inhibition of Na+,K+-ATPase during drug washout and because no consistent changes in myocardial sodium and potassium concentrations have been observed at the time of inotropic response. Four groups of experiments were conducted to test the hypothesis that cardiac Na+,K+-ATPase is the positive inotropic receptor for David Dalee Ku cardiac glycosides and that alterations in transmembrane sodium movement cause an increased myocardial contractility. First, rela- tionships among positive inotropic responses, Na+,K+-ATPase and sodium pump activities were studied using paced Langendorff preparations of guinea pig hearts. The active, uninhibited Na+,K+-ATPase activity was estimated from the initial velocity of (3H)-ouabain binding in ventri- cular homogenates, and sodium pump activity from ouabain-sensitive 86Rb uptake of ventricular slices. Development and loss of the ino- tropic response during ouabain or digitoxin perfusion and washout were accompanied by reduction and subsequent recovery of the initial velocity of (3H)-ouabain binding, respectively. Despite differences in the time course of the loss of inotropic responses to ouabain or digitoxin, the quantitative relationships between the Na+,K+-ATPase inhibition and inotropic response were similar. Inotropic responses to digitoxin during perfusion, and subsequent loss during washout, also were accom- panied by a reduction and subsequent recovery of the sodium pump activity. Thus, it appears that the inhibition of cardiac Na+,K+- ATPase by inotropic concentration of digitalis actually results in a significant inhibition of the sodium pump activity; furthermore, the magnitude of inotropic response and that of sodium pump inhibition are related. If Na+,K+-ATPase inhibition is causally related to the positive inotropic effects of the digitalis, the enzyme inhibition should cause a positive inotropic response regardless of the mechanism of such an inhibition. Therefore, in a second group of experiments, the rela- tionship between Na+.K+-ATPase inhibition by monovalent cations and David Dalee Ku their inotropic effect was studied in guinea pig heart. Rb+ (2-5 mM) or Tl+ (0.5-2.0 mM) inhibited Na+,K+-ATPase activity in the presence of Na+, K+ and MgZ+, whereas other monovalent cations such as K+, Cs+, NH4+, Na+ or Li+ (up to l0 mM) produced essentially no effect on the enzyme activity or slightly stimulated it. In left atrial strips stimulated with field-electrodes and bathed in Krebs-Henseleit solu- tion, only those monovalent cations capable of inhibiting Na+,K+- ATPase activity (Rb+ and Tl+) produced a sustained positive inotropic effect in the presence of propranolol. Positive inotropic effect of Tl+ was accompanied by a reduction in sodium pump activity. High concentrations of Li+ (20-30 mM), which are capable of inhibiting sodium pump activity, produced a sustained positive inotropic response. Thus, among monovalent cations, only sodium pump inhibitors produce a sustained positive inotropic response. The goal of the third group of experiments was to compare the effect of ouabain and monovalent cations on cardiac Na+,K+-ATPase and contractile force in guinea pigs and rats. Marked species-dependent differences were observed in the sensitivity of cardiac Na+,K+-ATPase to the inhibitory effect of ouabain. Equally marked species differences were observed in the concentration of ouabain needed to produce posi- tive inotropic effects in atrial preparations. In contrast, there were only minimal differences in these species in sensitivity of cardiac Na+,KA-ATPase to the inhibitory effect of Rb+ or Tl+. Simi- larly, there were also no remarkable differences in the sensitivity of rat and guinea pig atrial preparations to the positive inotropic action of Rb+ or Tl+. Thus, the positive inotropic response of David Dalee Ku isolated atrial preparations with various agents (ouabain, Rb+, and Tl+) appear to be related to the response of cardiac Na+,K+-ATPase to these inhibitors. In the final group of experiments, the relationship between altered transmembrane sodium movements and myocardial contractility was further studied by opposing the action of the sodium pump with grayanotoxins (GTX), agents previously shown to increase resting sodium influx. In electrically-driven guinea pig left atrial pre- parations, l uM GTX produced a slight depolarization and appeared to decrease the upstroke velocity of the action potential. GTX (0.5-5 uM) also produced a dose-dependent positive inotropic response in the presence of propranolol. At higher concentrations, GTX produced arrhythmias. Both positive inotropic and arrhythmic effects of GTX were reversible after the washout of the drug. Although GTX had no effect on partially purified Na+,K+-ATPase, they produced dose-dependent increases in ouabain—sensitive 86Rb uptake by ventricular slices under the condition in which intracellular sodium concentration determines the rate of active cation transport by Na+,K+-ATPase system. These data suggest that positive inotropic effects of grayanotoxin are due to an increased sodium permeability of the cytoplasmic membrane. Therefore, it is concluded, from the data obtained in these studies, that alterations of transmembrane sodium movements from either sodium pump inhibition by digitalis and monovalent cations + (Rb+, Tl and Li+) or opposing the sodium pump by grayanotoxins cause an increase in myocardial contractility. MECHANISM OF INOTROPIC ACTION OF CARDIAC GLYCOSIDES By David Dalee Ku A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology 1976 I would like to dedicate this dissertation to my mother, Yvonne. Her love and constant encouragement had helped make this possible. ii ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Tai Akera, for his interest in my research and his unfailing patience and guidance throughout the course of this work. In particular, I have appreciated his constant willingness to advise me regarding all aspects of a research career. This has enabled me to begin to develop as an independent researcher. I would also like to thank Dr. Theodore M. Brody for his interest and helpful suggestions in my research. I appreciate the concern of the members of my committee, Dr. Janice L. Stickney, Dr. Thomas Tobin and Dr. Henry w. Overbeck for their constructive assistance. Finally, I would like to thank Mrs. Ven-Jim (Kathy) Cheng, Mr. David McConnell, and Miss Kathy Toohey for their competent technical assistance. iii TABLE OF CONTENTS Page INTRODUCTION .............................................. l A. General Background .............................. l B. The Characteristics of the Sodium Pump+ .......... 2 C. Nature of the Cardiac Glycosides- Na+ ,K+ -ATPase Interaction ..................................... 5 D. Relationship between Na+,K+-ATPase Inhibition and the Positive Inotropic Effects of Cardiac Glycosides ...................................... 9 E. Specific Objectives ............................. 13 MATERIALS AND METHODS ..................................... 15 A. Materials ....................................... 15 B. Isolated Heart Studies .......................... l6 C. Transmembrane Potential Studies ................. 20 D. Cardiac Na+.K+-ATPase Preparations .............. 21 E. K-ATPase Studies ........................... 22 F. Cardiac Homogenate Preparations ................. 23 G. -Ouabain Binding and Release Studies ........ 24 H. Sodium Pump Studies ............................. 25 1. Miscellaneous Methods ........................... 27 J. Statistical Analysis ............................ 27 RESULTS ................................................... 28 A. The Relationship Among Na+,K+-ATPase, Sodium Pump Activity and the Positive Inotropic Response to Cardiac Glycosides ........................... 23 Isolated Heart Studies ..................... 28 (3H)-Ouabain Binding Studies ............... 3] 86Rb-Uptake Studies ........................ 39 B. The Relationship Between the 0uabain-Na+,K+- ATPase Complex Formed in the Functioning Heart and Formed jfl_vitro in the Presence of Various Ligands ......................................... 41 iv Page C. The Relationship Between Sodium Pump Inhibition and Myocardial Contractile Force ................ 48 Na+,K+-ATPase Activity ..................... 49 Isolated Atria Studies ..................... 53 D. Comparison of the Effect of Monovalent Cations and Ouabain on Cardiac Na+,K+-ATPase and Contractile Force ........................................... 75 Isolated Atria Studies ..................... 78 Na+,K+-ATPase Studies ...................... 84 E. Relationship Between the Alterations of Sodium Ion Movements and Myocardial Contractile Force.. 94 Effects of Grayanotoxins on Electromechani- cal Properties of Guinea Pig Left Atrium... 96 Effects of Grayanotoxins on Contractile Force ...................................... 98 Effects of Grayanotoxins on Na+,K+-ATPase Activity ................................... l08 Effects of Grayanotoxins on Sodium Pump Activity ................................... llO DISCUSSION ................................................ 114 A. The Relationships Between Cardiac Na+,K+-ATPase Inhibition and Myocardial Contractile Force With Cardiac Glycosides .............................. ll5 B. Effects of Monovalent Cations on Cardiac Na+,K+- ATPase Activity and Myocardial Contractile Force ll8 C. Comparison of the Effect of Cardiac Glycoside and Monovalent Cations on Cardiac Na+,K+-ATPase and Contractile Force in Guinea Pig and Rat Hearts .......................................... l23 D. The Relationship Between Sodium Pump Inhibition and Myocardial Contractile Force ................ 129 E. The Relationship Between the Alterations of Sodium Ion Movements and Myocardial Contractile Force ........................................... 132 Page SUMMARY AND CONCLUSIONS ................................... 139 APPENDIX .................................................. l4O BIBLIOGRAPHY .............................................. T42 vi LIST OF TABLES Table Page l Effects of monovalent cations on cardiac Na+, K+ -ATPase activity assayed in the presence of 150 mM Na+ and 5. 8 mM K+ ...................... 50 2 Effects of monovalent cations on cardiac Na+ ,K +-ATPase activity assayed in the presence of 20 mM Na+ and 5.8 mM K+ .................................... 52 3 Effects of thallous ion on cardiac Na+,K+-ATPase activity ......................................... 54 4 Effect of thallous ions on ouabain-sensitive 86Rb uptake and intracellular thallous ion concentra- tion in guinea pig hearts ........................ 7O 5 Effect of Li+ on ouabain-sensitive 86Rb uptake and intracellular Li+ concentration in guinea pig hearts ........................................... 76 6 Effects of grayanotoxins on electromechanical properties of guinea pig atria in the presence of propranolol ................................... 97 7 Effect of grayanotoxin I and a-dihydro- grayano- toxin II on Na+, K+ -ATPase activity ............... 109 8 Effect of grayanotoxins on ouabain-sensitive 86Rb uptake (without sodium loading) .................. lll 9 Effect of grayanotoxins on ouabain-sensitive 86Rb uptake (with sodium loading) ..................... ll3 vii Figure 10 ll LIST OF FIGURES Page The onset and duration of the positive inotro- pic action of ouabain and digitoxin in elec- trically driven isolated guinea pig hearts ...... 29 The effect of ATP on the binding of (3H)-ouabain to homogenates obtained from isolated perfused guinea pig hearts ............................... 32 The ATP-dependent (3H)-ouabain binding of homo- genates obtained from control and ouabain- perfused guinea pig hearts ...................... 34 Effects of ouabain on cardiac contractile force and on initial (3H)-ouabain binding velocity.... 35 Effects of digitoxin 03 cardiac contractile force and on initial ( H)-ouabain binding velocity ........................................ 38 Effects of digitoxin on cardiac gontractile force and on ouabain-sensitive 8 Rb uptake ...... 40 Dissociation of (3H)-ouabain from dog heart homogenate ...................................... 43 Dissociation of (3H)-ouabain from ventricular tissues after puppy heart perfusion of the drug. 47 Effects of Cs+ and NH + on electrically-driven guinea pig atrial strips in Krebs-Henseleit solution ........................................ 55 Effects of K+, Rb+ and Li+ on electrically-driven guinea pig atrial strips in Krebs—Henseleit solution ........................................ 57 The positive inotropic effect of Rb+ on isolated, electrically-driven left atria of guinea pig heart as a function of time and the number of contractions .................................... 50 viii Figure 12 l3 T4 15 l6 l7 l8 l9 20 21 22 23 24 Effect of Tl+ on electrically-driven guinea pig atrial strips in Krebs-Henseleit solution... Effect of Tl+ on isolated, electrically-driven left atria of guinea pig heart when exposure to the drug was restricted to non-contracting tissue .......................................... A typical tracing of the effect of Li+ on iso- lated left atria of guinea pig heart in Krebs- Henseleit solution .............................. Typical tracing of the effects of sucrose— and Li -substituted Krebs-Henseleit solution on iso- lated, electrically-driven guinea pig left atria strips .................................... Ouabain effect on isometric contractile force of guinea pig left atrial preparation .............. Ouabain effect on isometric contractile force of rat left atrial preparation ..................... Dose-response relationship of the positive ino— tropic effect of ouabain in guinea pig and rat left atrial preparations ........................ Dose- -response relationship of the positive ino- tropic effect of Rb+ in guinea pig and rat left atrial preparations ............................. Dose- -response relationship of the positive ino- tropic effect of Tl+ in guinea pig and rat left atrial preparations ............................. Inhibition of guinea pig and rat heart Na+,K+- ATPase by ouabain ............................... Inhibition of guinea pig and rat heart Na+ ,K+ — ATPase by Rb+ ................................... Inhibition of guinea pig and rat heart Na+ ,K+ - ATPase by Tl+ ................................... Dissociation of ( 3H)-ouabain from homogenates obtained from various areas of guinea pig hearts .......................................... ix Page 62 66 71 74 79 8O 82 83 85 86 88 89 91 Figure 25 26 27 28 29 3O 31 Effect of ouabain on 86Rb uptake of ventricular slices obtained from guinea pig and rat hearts.. Log dose-response curves for the positive ino- tropic effect of grayanotoxins in the presence and absence of propranolol ...................... Rates of development of positive inotropic responses to grayanotoxins in guinea pig atrial preparations .................................... Rates of development of positive inotropic responses to grayanotoxins in the presence of propranolol ..................................... Nashout of positive inotropic effect of grayano- toxins in electrically-driven guinea pig left atrial preparations ............................. Nashout of positive inotropic effect of grayano- toxins in the presence of propranolol ........... Typical tracings of the effects of grayanotoxins on the contractile force of guinea pig left atria ........................................... Page 95 99 TO] 103 104 105 106 INTRODUCTION A. General Background Digitalis is commonly referred to the entire group of cardiac glycosides and aglycones and can be obtained from digitalis leaf, stro- phanthus, and squill. Digitalis has been used in the treatment of congestive heart failure for nearly two hundred years. Despite numerous studies, the molecular mechanism of the action of digitalis remains unknown (see Lee and Klaus, l97l). The pharmacologic action of digitalis, i.e., its direct effect on myocardial contractility has been firmly established only since the classic experiments of Cattell and Gold in 1938 which showed that digitalis increases the force of contraction of isolated, electrically-driven cat papillary muscle and thus demonstrated that digitalis acts directly on heart muscle to augment its contractility. Prior to that time, it was believed that these agents improved the impaired systemic circulation by one of a number of extracardiac actions: through a diuretic action on the kidneys, by slowing of the heart rate through a vagal effect, or by reducing venous return through a relaxing action on the tone of the systemic venous bed. In l953, Schatzmann observed that the cardiac glycosides speci- fically inhibit the active transport of sodium and potassium across the cell membrane of the human erythrocyte. Na+,K+-ATPase (Mg2+- dependent, (Na+,K+)-activated ATP phosphohydrolase, EC 3.6.l.3.), l 2 which is responsible for the active transport of sodium and potassium ions across the cytoplasmic membrane, was subsequently shown to be highly sensitive to the inhibitory effects of the cardiac glycosides (Skou, l957; Glynn, 1964). Therefore, it has been speculated that alterations of the intracellular ionic environment, namely changes in intracellular sodium and potassium concentrations, may be causally related to the positive inotropic effects of digitalis (Hajdu and Leonard, l959). A causal relationship between the Na+,K+-ATPase inhibition and the positive inotropic responses of cardiac glycosides, however, has never been firmly established; however, the Na+,K+- ATPase or the sodium pump system is the only biochemical event which is consistently and specifically affected by pharmacologic concen- trations of digitalis. B. The characteristics of the sodium pump Cardiac muscle cells at rest, like other muscle and nerve cells, maintain high intracellular potassium and low intracellular sodium ions concentrations and have a transmembrane voltage between 80 and 90 mV negative to the extracellular fluid potential. There is a continuous passive influx of sodium and efflux of potassium following chemical gradients across the cell membrane. The passive fluxes of ions, however, are balanced by an active, coupled efflux and influx of sodium and potassium ions, respectively, such that the net flux of all ions during the steady state is zero. The mechanism responsible for the active transport of sodium out of the cell and potassium into the 3 cell is a sodium-potassium pump or commonly referred to as the "sodium pump". This sodium pump is fundamental to many physiological processes and required for maintenance of excitability. Most of the studies on the nature of the intact cation transport system were performed using resealed erythrocyte ghosts (Hoffman, l958; Whittam, l962) and the internally perfused giant squid axon (Caldwell gt__l,, l960) primarily because of the simplicity and good reproducibility of these preparations. Results of these studies have revealed that the sodium pump is asymmetrically oriented within the membrane. The outward direction of the reaction cycle, for example, requires the presence of internal sodium and ATP and external potassium. Furthermore, it was demonstrated that the intrinsic rate of sodium pump activity is directly proportional to the intracellular sodium concen- tration. When the intracellular sodium concentration is low, for example, during the resting state, the intracellular sodium concen- tration is the limiting factor for sodium pump activity. The asymmetry of the sodium pump, in part, is believed to be the basis of the ability of the pump to respond to fluctuating sodium influxes during the phases of each cycle of cardiac contraction. The movement of sodium and potassium ions is coupled. The stoichio— metry of this transport system is believed to be three sodium ions pumped out per two potassium ions pumped in (Garahan and Glynn, l967). The energy necessary for these cations transport is supplied by the hydrolysis of ATP (Hoffman, 1960) and an adequate supply of ATP depends on the metabolic processes in the cell. Application of ATP by 4 micropipet to the inside of the membrane increases transport while application to the outside has no effect (Caldwell et_al,, l960). These data suggest that the active site for ATP resides on the internal surface of the membrane. Metabolic poisons which prevent the formation of ATP can also inhibit this transport mechanism (Hodgkin and Keynes, 1955). In addition, it was shown that inosine triphosphate (ITP) supports cation transport less effectively and other nucleotides are ineffective (Hoffman, l960). Therefore, if an enzyme is responsible for converting chemical energy to mechanical energy for these active cations transport, its activity must specifically require ATP. Skou (1957) isolated an enzyme from the peripheral nerve of the crab which hydrolyzed ATP into ADP and Pi and was activated by Mg2+. The enzyme could be further stimulated several fold by the simul- taneous addition of sodium and potassium. Since this enzyme system was obtained from fragmented membrane preparations, it was postulated that this MgZ+-activated, Na+ and K+-dependent ATP phosphohydrolase (ATPase) could be responsible for the active transport of Na+ and K+ across cell membranes. Later studies on erythrocyte membranes by several groups established that the sodium pump and Na+,K+-ATPase share many features in common (Glynn, l964,.Skou, 1965). The simi- larities of these fragmented membrane preparations compared to those from the intact transport system strongly suggested that Na+,K+- ATPase is the enzymatic expression of the sodium pump. The characteristics of asymmetry of the intact transport system were also found in the enzyme-containing membranes with respect to its response to Na+ and K+. The K+ ion stimulated the 5 enzyme activity only at the outer surface of the membrane whereas Na+ was effective only at the inner surface (Baker, 1963). A variety of other monovalent cations such as Rb+, Tl+, Cs+, NH4+ and Li+ is capable of substituting for K+ at the potassium-activation sites, but no other monovalent cations, at the present, have been shown to be capab1e of effectively substituting for Na+ at the sodium-activation sites (Skou, 1960, 1965). Skou (1957) has shown that the Na+/K+ ratio, rather than the concentrations of individual cations is the primary determinant of enzyme activity. Thus, increasing the concen- tration of one cation without increasing the other cation concurrently could cause an inhibition of enzyme activity. This inhibition of enzyme activity, for example, in a high potassium concentration and low sodium concentration medium, is thought to be the result of dis- placement of Na+ by K+ for the sodium-activation sites (Skou, 1957). C. The nature of the cardiac glycosides-Na+,K+-ATPase interactions Schatzmann (1953), in studying sodium and potassium fluxes in the red blood cell, found that cardiac glycosides specifically inhibit the active transport of Na+ and K+ across the membrane. Cardiac glyco- sides apparently inhibit the sodium transport system only when they were applied externally but not when injected directly into the intracellular space (Caldwell and Keynes, 1959). This suggests that the receptor for the drug resides on the external surface of the membrane (Hoffman, 1966; Whittam, 1962). Since the active transport of cations requires ATP and the active site for ATP apparently resides 6 on the internal surface of the membrane, it was concluded that inhi- bition by the cardiac glycoside is an allosteric event. In other words, alterations in the conformation of the pump must be invoked to connect glycoside interaction on the external surface to inhibition of ATP hydrolysis at the internal surface. In l968, Matsui and Schwartz quantitatively measured the inter- action of cardiac glycosides with Na+,K+-ATPase. They examined 3H- digoxin binding to fragmented membrane preparations isolated from beef heart that contained Na+,K+-ATPase activity. The enzyme system bound glycosides most rapidly and in greatest quantity when they were incu- 2+ and Na+ or in the presence of Mg2+ bated in the presence of ATP, Mg and Pi. The amount of drug bound to these preparations correlated well with Na+,K+-ATPase activity (i.e., higher the activity, the greater the binding). The specificity of this binding was evident by the fact that only the cardioactive glycosides were bound to the enzyme. The inactive or less active cardiac glycosides (with respect to their effects on cardiac muscle) and other steroids could not compete with digoxin, ouabain, or other cardioactive glycosides for the receptor. Subsequently, Hansen gt_al, (1971) demonstrated that the binding of cardiac glycosides to Na+,K+-ATPase results in a stoichiometric inhibition of enzyme activity. Further, the release of cardiac glycosides from the enzyme results in a reactivation of the enzyme activity (Akera and Brody, 1971). If the bound drug was actually responsible for the enzyme inhi- bition (as well as the cardiotonic effect) the characteristics of the 7 1n vivo-formed complex should be identical with those of a Na+,K+- ATPase-glycoside complex formed in_vit§g, The glycoside-Na+,K+- ATPase complexes formed jg_vjtrg_seem to have different properties dependent upon the ligand conditions which prevail during the binding reaction. Such differences may be detected by monitoring the disso- ciation reaction of the ouabain-enzyme complex in a mixture different from the binding medium and having a low ionic strength (Akera and Brody, 1971). It appears that there are at least three different forms of the ouabain-enzyme complex prepared ig_vjtrg_with rat brain enzyme (Akera et_al,, 1976). The ouabain-enzyme complex formed with 2+ and Pi was stable and unaffected by K+ added to the dissociation 2+ M9 medium. The ouabain-enzyme complex formed in the presence of Mg and ATP or Na+, K+, M92+ and ATP had an intermediary dissociation rate in the absence of K+ and was not stabilized by K+. The addition of Na+ to the binding medium containing Mg2+ and ATP increased the ouabain binding and resulted in the formation of an unstable complex. The addition of K+ to the dissociation medium stabilized the complex. Thus, it would be interesting to determine if similar ouabain-enzyme complexes can be formed with the cardiac enzyme and to determine which form of the ouabain-enzyme complex is relevant for the study of drug- enzyme interactions occurring in the functioning heart. Addition of potassium to the binding medium containing Na+, Mg2+ and ATP retarded the rate of binding of cardiac glycosides to Na+,K+- ATPase (Matsui and Schwartz, 1968; Schwartz gt_al,, 1968; Allen gt_al,, 1970; Akera and Brody, 1971). That potassium retards and sodium 8 accelerates the development of the inotropic response to cardiac glycosides was also observed in isolated myocardial preparations (Caprio and Farah, 1967; Prindle gt_gl,, 1971). Since the potassium- activation sites are located on the external surface of the membrane system and cardiac glycosides appear to interact with the pump only when the drug is in the external medium, this potassium antagonism was originally thought to reflect a competition between the glycosides and potassium for potassium-activation sites (Glynn, 1957). However, further studies indicated that the amount of ouabain bound in the presence of a combination of Na+. K+. Mg2+ and ATP over a long period equals the amount bound in the presence of only Na+, Mg2+ and ATP (Allen and Schwartz, 1970). Similarly, Prindle et_al, (1971) have demonstrated, in the isolated cat papillary muscle, that the rate of development of the inotropic response to the cardiac glycosides was inversely related to the extracellular potassium concentration, but that the peak contractile response to cardiac glycosides was unaltered by the potassium concentration. The results of these studies indicate that the effect of potassium on the glycoside-Na+,K+-ATPase interaction was apparently to decrease the rate of complex formation between the drug and the receptor and that potassium is not a competitive inhibitor of the.glycoside-enzyme interaction. Similarly, the sodium-dependence of the rate of development of the positive inotropic action of cardiac glycosides on heart muscle (Caprio and Farah, 1967) resembles the sodium-dependence of the 9 inhibitory effect of cardiac glycosides on isolated Na+,K+-ATPase. It has been demonstrated repeatedly that the inhibitory action of cardiac glycosides on membrane ATPase is facilitated by increasing and reduced by lowering the sodium concentration (Repke and Portius, 1963; Repke, 1964; Schwartz gt_al., 1968). D. The relationship between Na+,K+-ATPase inhibition and the positive inotropic effects of cardiac glycosides In 1963, Repke suggested the possibility that Na+,K+-ATPase may be the pharmacologic receptor for the cardiac glycosides. He showed that a positive correlation exists between the sensitivity of the heart to inotropic effects of cardiac glycosides and the sensitivity of Na+,K+-ATPase to the jg_vjtrg inhibitory effect of the cardiac glycosides in various animal species (Repke gt_al,, 1965). In the same species, the potency of different cardiac glycoside preparations to inhibit Na+,K+-ATPase jn_vjtrg_parallels that necessary to produce the inotropic effects (Repke, 1964; Repke gt_al,, 1965). One of the remarkable features of the positive inotropic effect of cardiac glycosides is the divergent species—dependent difference in sen- sitivity to the glycosides. Although part of such species effects may be explained by variations in the rate of metabolism and elimination of the drug (Russel and Klaassen, 1972), similar observations of a low myocardial sensitivity to cardiac glycosides have also been con- firmed in isolated rat heart preparations (Masuoka and Saunders, 1950; Koch-Neser and Blinks, 1962). Thus, a difference in metabolism of cardiac glycosides could not account for this finding. In attempts to explain the relative insensitivity of the rat myocardium to digitalis, 10 Tobin and Brody (1972) have suggested that the differences in sensi- tivity to or affinity for Na+,K+-ATPase to cardiac glycosides are due to differences in dissociation rate constants of the drug-enzyme complex, whereas the association rate constants are rather similar. Thus, Na+,K+-ATPase obtained from digitalis-sensitive species, such as human, dog or cat have a high affinity for digitalis, whereas those obtained from digitalis-insensitive species, such as the rat and the toad, have a low affinity. Other species which are moderately sensitive to digitalis have cardiac Na+,K+-ATPases which have intermediate affinity (Repke et_al,, 1965; Akera et_al,, 1969; Allen and Schwartz, 1969). The relationship between cardiac glycoside-induced inhibition of Na+,K+-ATPase and species sensitivity was further studied by Akera gt_ a1, (1973). These investigators showed in several mammalian species that the release of (3H)-ouabain previously bound to cardiac Na+,K+- ATPase jn_vjtrg_has a similar time constant as the loss of the ino- tropic response to ouabain in Langendorff preparations in several species. The rates of onset of inotropism and the jn_vitro ouabain binding to Na+,KI-ATPase are quite similar in all species, despite their diverse sensitivity to the drug. The studies further support the contention that dissociation rates (i.e., as opposed to association rates) are related to species sensitivity to the glycosides as reported by Tobin and Brody (1972). From these parallelisms between the inhibitory action on Na+,K+- ATPase jn_yjt§g_and the inotropic action of cardiac glycosides it was suggested that the inhibition of Na+,K+-ATPase is the mechanism of the 11 positive inotropic action of cardiac glycosides. However, the actual measurement of the degree of Na+,K+-ATPase inhibition following a pre- determined magnitude of inotropic response of cardiac glycoside has not been obtained. Akera §t_al, (1969) were first to demonstrate that cardiac Na+,K+-ATPase activity was indeed inhibited in the intact animal after ouabain administration. A mild to moderate (20 to 40%) inhibition of Na+,K+-ATPase was observed when this enzyme was isolated from dog heart exposed to positive inotropic doses of ouabain (Akera gt_al,. 1970). The inhibition of the Na+,K+-ATPase by cardiac glycosides jn_ situ_was also shown by Besch et_a1, (1970). In addition, neither mitochondria nor sarcoplasmic reticulum appeared to be affected by these inotropic doses of cardiac glycosides (Besch et_al,, 1970). However, some investigators disagree with the view that the .inhibition of Na+,K+-ATPase is directly related to the inotropic responses of cardiac glycosides (Lee and Klaus, 1971). Furthermore, in the canine jg_§jtu_experiments mentioned above, carried out by Akera gt_al, (1969, 1970) and Besch gt_al, (1970), a subsequent washout of the inotropic effect and concurrent recovery of cardiac Na+,K+-ATPase activity to normal level was not reported. Okita and his associates have suggested that the inhibition of the Na+,K+-ATPase might be associated with the toxic action of cardiac glycosides rather than with the therapeutic effects (Okita gt a1,, 1973; Ten Eick gt_al,, 1973). These investigators, in their attempt 12 to dissociate the inhibition of the enzyme activity from positive inotropism, employed an isolated perfused rabbit Langendorff pre- paration in which they were able to observe that the augmented con- tractile response returned to control levels after removal of ouabain. However, the initial inhibition of enzyme activity associated with the positive inotropic action remained at the same level after restora- tion of the augmented contractility to control levels. Consequently, these investigators suggested that the inotropic effect was not asso- ciated with changes in Na+,K+-ATPase activity. Thus, from these many divergent experimental findings, it is apparent that additional studies on the relationship between cardiac Na+,K+-ATPase inhibition and inotropic effects of cardiac glycosides, particularly during the time course of the development and the dissipa- tion of the inotropic response, are needed before a causal relation- ship can be established between these phenomena. If Na+,K+-ATPase inhibition is causally related to the positive inotropic effects of cardiac glycosides, the same relationship should exist between Na+,K+-ATPase inhibition and changes in cardiac con- tractile force regardless of the mechanism of enzyme inhibition. The erythrophleum alkaloid, cassaine, for example, shares many of the pharmacological actions of the cardiac glycosides, including a positive inotropic action, yet lacks the structural characteristics typical of cardiac glycosides. Cassaine also inhibits Na+,K+-ATPase by interacting at the cardiotonic steroid binding sites of the enzyme (Tobin gt_al,, 1975). The principal difference between the interaction 13 of cassaine and that of ouabain with Na+,K+-ATPase appeared to be the more rapid dissociation of cassaine from the cardiotonic steroid binding site(s) of Na+,K+-ATPase. Consistent with this observation, the rates of offset of cassaine-induced inotropism in Langendorff perfused dog and guinea pig hearts were also several times faster than those of ouabain-induced inotropism (Tobin et_al,, 1975). Other Na+,K+—ATPase inhibitors, such as p-chloromercuribenzoate, N-ethylmaleimide, ethacrynic acid and fluoride (Glynn, 1963; Skou and Hilberg, 1965; Duggan and N011, 1965; Opit gt_al., 1966) have been shown to produce positive inotropic effects in isolated hearts (From and Probstfield, 1971; Yamamoto gt_al,, 1973; Covin and Berman, 1959; From _t._1,, 1971; From, 1970). These agents, however, induce other biochemical effects as well and it is difficult to determine if the observed changes in contractile force resulted from cardiac Na+,K+- ATPase inhibition. Recently, it was demonstrated that Rb+ inhibits Na+,K+-ATPase in the presence of Na+ and k+ (Tobin et 11., 1974). Thus, studies with Rb+ and other monovalent cations with differential effects on Na+,K+- ATPase activity and myocardial contractile force would be useful in elucidating the relationship, or lack of relationship, between these two phenomena. E. Specific Objectives The present study was conducted to test the hypothesis that Na+,K+—ATPase is the pharmacologic receptor for the cardiac glyco- sides. In order to demonstrate that the alteration of the l4 intracellular ionic environment, namely changes in intracellular sodium concentrations, is causally related to the positive inotropic effects of digitalis, the following three groups of experiments were conducted. The first group of experiments was undertaken to further investi- gate the relationships among cardiac Na+,K+-ATPase inhibition, sodium pump activity and the positive inotropic responses to cardiac glyco- sides, particularly during the time course of the development and dissipation of the inotropic response. In addition, the characteristics of the ouabain-Na+,K+-ATPase complex formed following drug perfusion of a beating heart were compared with those of the drug-enzyme complexes formed jfl_vjtrg_in the presence of various ligands to determine if both complexes are similar and reversible as previously demonstrated (Akera §t_al,, 1976). The second group of experiments was designed to study the effects of various monovalent cations on cardiac Na+,K+-ATPase activity and cardiac contractile force and to determine if inhibition of cardiac Na+,K+-ATPase invariably resulted in an enhancement of myocardial contractility. In addition, the characteristics of monovalent cation- induced and cardiac glycoside-induced positive inotropic responses were compared to determine if both agents mediated their inotropic effect via a common mechanism. The final group of experiments was conducted to study the effect of an increased resting sodium influx by grayanotoxins on myocardial 15 contractility and to determine if the results are consistent with a hypothesis that altered transmembrane sodium movements affect myo- cardial contractility. MATERIALS AND METHODS A. Materials Ouabain (Strophanthin-G) (3H-1abelled with a specific radio- activity of 12 Ci/mmole), rubidium chloride (86Rb-labelled with average specific radioactivity of 0.2 Ci/mmole, and methoxy-inulin (methoxy- 3H-labelled) with specific radioactivity of 102.5 mCi/g) were all purchased from New England Nuclear, Boston, Mass. Ouabain octahydrate (strophanthin-G), adenosine-S'-triphosphate (disodium salt), Tris ATP [Tris(hydroxymethy1)—aminomethane-ATP], (i) propranolol hydrochloride, tetrodotoxin and digitoxin were purchased from Sigma Chemical Company, St. Louis, Mo. Sucrose, special enzyme grade, was obtained from Schwartz/Mann, Orangeburg, N.Y. Rubidium chloride, cesium chloride, thallous nitrate and thallous acetate (ultrapure grade) were purchased from Ventron Alfa Products, Beverly, Mass. Other chemicals were of analytical reagent grade. Grayanotoxin I and a-dihydro-grayanotoxin II were prepared by Dr. Junkichi Iwasa according to the method of Miyajima and Takei (1934). B. Isolated Hearts Studies The isolated perfused heart (Langendorff) preparation was employed since it possesses many properties that are similar physiologically and metabolically to those of the intact working heart (Dobson, l6 17 gt_§l,, 1974; Bunger _t_al,, 1975). The direct perfusion of the muscle cells via the vascular beds enables one to study a variety of myocardial functions whenever precise control of mechanical perfor- mance, metabolism and extracellular cations is required. In the present study, the guinea pig was selected as the major experimental animal species primarily because it is intermediately sensitive to digitalis and for other practical reasons, such as size, availability and cost. In addition, the time course of both the development and dissipation of the positive inotropic response are much easier to carry out and to control using the guinea pig Langendorff prepara- tion. However, other animal species, such as the puppy (digitalis- sensitive) and rat (digitalis-insensitive), were also used for comparative studies. Animals unselected as to sex were anesthetized; their hearts were immediately excised and cannulated via the aortic root on a modified Langendorff perfusion apparatus (Akera gt_al,, 1973). The hearts were perfused with a constant flow rate, by means of a roller pump, of Krebs-Henseleit solution of the following millimolar composition: NaCl, 118.0; NaHCO3, 27.2; KC1, 4.8; KH2P04, 1.0; MgSO4, 1.2; CaClz, 2.5; glucose, 11.1 (Ninegrad and Shanes, 1962). The Krebs-Henseleit solution was saturated with a 95% 02-5% C02 mixture (pH 7.4) and maintained at a temperature of 30°C. Temperature was monitored just above the inlet cannula to the heart. 18 When the perfusate coming from the preparation was clear and the heart had a regular rhythm, both atria were removed and the heart was electrically-driven by an electrode placed at or close to the A-V node with 3 msec pulses from a Grass S44 stimulator at 1.5 Hz and a voltage of 10% above threshold. Isometric contractile force was continuously monitored using a force-displacement transducer (Grass, FT-O3C) attached to the apex of the heart with a resting tension of 2 g. Heart electrograms, perfusion pressure, perfusion fluid temperature and the first derivative of the force of contraction were also continuously monitored in all experiments. A 45-min equilibration period was allowed for each heart before the start of the experiment. Following equilibration, either 1.2 uM ouabain or 0.4 uM digitoxin in Krebs-Henseleit solution was perfused for 20 min, at which time a substantial positive inotropic response was obtained. For the series of drug "washout" experiments, drug-free solution alone was continued after the termination of either the ouabain or digitoxin perfusion, and the decay of the inotropic effect was monitored. Langendorff preparations that served as controls for the drug-treated experiments were perfused with Krebs-Henseleit solu- tion for times comparable to drug-treated hearts. At appropriate times during the development and dissipation of the inotropic re- sponses of cardiac glycosides, the residual active, uninhibited portion of Na+,K+-ATPase activity was estimated from the initial velocity of ATP-dependent (3H)-ouabain binding (for details, see below). 19 In one series of experiments, (3H)-ouabain instead of the un- labelled ouabain was perfused into the puppy Langendorff preparations. After 20 min drug perfusion, the ventricular muscle was excised, homogenized, centrifuged at 100,000 x g for 30 min at 0°C. The pellet was subsequently resuspended in 10 mM Tris-HCl buffer (pH 7.5) and the release of (3H)-ouabain from the resuspended homogenate was monitored at 37°C to determine the dissociation rate constant and the effect of KCl and NaCl on this rate constant. For the studies of the effects of monovalent cations and grayano- toxins on myocardial contractile force, a field-electrode stimulated guinea pig (350-500 9, either sex) and Sprague-Dawley rat (325-375 9, male) atrial preparations were used. Since RbCl and KC1 produced a negative inotropic response with point stimulation (as in Langendorff preparations), presumably due to the effect of these cations on impulse propagation, the use of field-electrode stimulation was essential in these experiments. Briefly, the left atrium was dissected free from the Langendorff preparation and trimmed into one fan-shaped layer and mounted on a fixed electrode-clamp apparatus as described by Levy (1971). The temperature of the bathing medium was maintained at 30:0.1°C. The atrial strips were electrically stimulated with two platinum field-electrodes at 1 Hz, unless otherwise indicated, with square-wave pulses of 3 msec and a voltage not exceeding 15% above threshold. Isometric contractile force was recorded with a FT-O3C force-displace- ment transducer and a Grass polygraph. Resting tension was adjusted 20 to 0.8-1.0 g. A 60 min equilibration period was allowed for each atrial preparation before the start of the experiment. Following equilibration, the effect of addition of cardiac glycoside or other agents (monovalent cations and grayanotoxins) was studied. Changes in isometric contractile force were expressed as a percentage of the contractile force observed before the addition of the drug. In some instances, drug-induced changes in isometric contractile force were also normalized on a basis of percent of maximal change and plotted against either the number of contractions or time. Contractile force of control preparations was generally unchanged during the 30-min experimental period. For the series of drug "washout" experiments, five minutes after the maximal inotropic response was reached, the medium was washed twice with drug-free, pre-warmed and pre-oxygenated Krebs—Henseleit solutions, The dissipation of the inotropic response was monitored for a predetermined period of time or until the contractile force re- turned to control levels. In one series of experiments, quiescent studies were performed by turning the stimulator off. Details of the use of "quiescent periods" are described in Results. C. Transmembrane Potential Studies Guinea pig (350-500 9) left atrium was isolated and trimmed as described above, and suspended horizontally in a chamber containing oxygenated Krebs-Henseleit solution maintained at 30:0.l°C. Intra- cellular recordings of electrical potential were determined using 21 3 M KC1-filled glass microelectrodes, possessing resistances of 10-30 megohms, mounted at the end of a flexible tungsten wire and connected to the input of a Medistor A35 electromotor amplifier. The atrium was electrically stimulated and isometric contractions were recorded as described above. Action potential and isometric contractions were displayed on a Tektronix 502A oscilloscope and recorded with a Grass Model C4 oscilloscope camera. Tracings were measured for the amplitude of action potential, resting tension and the peak tension. A 45—min equilibration at 1 Hz stimulation was allowed for each atrial preparation before the start of the experiment. Following equilibration, the effect of drug addition was studied in each pre- paration, and the changes in isometric contractile force and action potential parameters were determined. 0. Cardiac Na+,K+-ATPase Preparations Cardiac Na+,K+-ATPase preparations were obtained from ventricular muscle of guinea pigs (either sex, 350-500 g) or Sprague-Dawley rats (male, 325-375 g) as previously described (Akera gt_al,, 1969) with minor modifications. Briefly, ten grams of ventricular muscle (ten to twelve guinea pig or rat hearts were pooled) were homogenized, with a Dounce ball-type homogenizer in 4 volumes of a solution containing 0.25 M sucrose, 5 mM histidine, 5 mM disodium ethylenediamine tetra- acetate (EDTA), 0.15% sodium deoxycholate and 0.01 mM dithiothreitol (pH adjusted to 6.8 with Tris-base). Deoxycholic acid and dithio- threitol were added immediately before use. The homogenate was centri- fuged at 10,000 x g for 30 minutes. The resulting supernatant was 22 centrifuged for 60 minutes at 100,000 x g. The residue was suspended and recentrifuged at the same speed for 30 minutes, and the final pellet was suspended in 10 m1 of a solution containing 0.25 M sucrose, 5 mM histidine and 1 mM Tris-EDTA and was added to an equal volume of 2.0 M NaI solution and then stirred for one hour. The mixture was diluted with 2.5 volumes of 1 mM Na2 EDTA and centrifuged twice for 60 and 30 minutes at 100,000 x g, resuspending each time with the sucrose-histidine-EDTA suspending solution. All procedures were carried out at 2°C. The final resuspension was stored frozen until use. For the study of the effect of grayanotoxins on Na+,K+-ATPase activity, rat brain enzyme preparations were used initially because they have the highest specific activity among various Na+,K+-ATPase preparations obtainable in this laboratory. Similar studies, however, were also performed with guinea pig heart enzyme since the data should be relevant for cardiac studies. Brain Na+,K+-ATPase preparations were obtained from Sprague-Dawley rats (male, 200-300 g) by the method of Akera gt_al, (1976). E. Na+,K+-ATPase Studies ATPase activity of the preparation was assayed by measuring the inorganic phosphate liberated from ATP (Akera gt_a1,, 1969). The incubation medium contained enzyme preparations (80-120 ug of protein per milliliter) in a 1.0 ml medium containing 5 mM Tris-ATP, 5 mM MgCl2 and 50 mM Tris-HCl buffer (pH 7.5) in the presence and absence of NaCl and KC1. MgZ+-ATPase activity assayed in the absence of 23 NaCl and KC1 was subtracted from total ATPase activity assayed in the presence of NaCl and KCl to calculate Na+,K+-ATPase activity. Various concentrations of each monovalent cation' or grayanotoxin were added to the incubation mixture. After a 5-min period the ATPase reaction was started by the addition of ATP and terminated 15-min later by the addition of 1.0 ml of ice-cold 0.8 N perchloric acid. For the study of ouabain inhibition of Na+,K+-ATPase, this procedure results in an underestimation of the enzyme inhibition (Akera, 1971). Therefore, in the glycosides experiments, enzyme preparations were incubated at 37°C for 10 min with various concentrations of ouabain in the presence of 100 mM NaCl, 5 mM MgC12, 5 mM Tris-ATP and 50 mM Tris—HCl buffer (pH 7.5). The Na+,K+-ATPase reaction was started by the addition of KC1 (final concentration, 15 mM) to the mixture. Values obtained with the addition of water were subtracted from those obtained with the addition of KC1 to calculate the ATPase activity stimulated by the simultaneous presence of Na+ and K+. Drug effects were expressed as percent inhi- bition of the Na+,K+-ATPase activity. Control Na+,K+-ATPase activity of the same enzyme preparation assayed concurrently in the absence of the drug but under the same experimental conditions was set at 100%. F. Cardiac Homogenate Preparations Cardiac homogenate preparations were obtained from mongrel puppies (500-1000 g) and guinea pigs (350-500 g) of either sex. A 0.5 9 portion of the left ventricular muscle was minced and homo- genized in 9 volumes of 0.3214 sucrose, using a Dounce ball-type 24 homogenizer and a motor-driven Potter-Elvehjem homogenizer. All procedures were carried out at 2°C. G. (3H)-0uabain Binding and Release Studies The initial velocity of (3H)-ouabain binding was determined by incubating the homogenate (4 mg of tissue per milliliter) with 0.01 uM (3H)-ouabain in the presence of 5 mM MgC12, 200 mM NaCl, 5 mM Tris- ATP, and 50 mM Tris-HCl buffer (pH 7.5) at 37°C for 1, 3, 5, or 10 min. Since preliminary experiments indicated that higher incubation temperatures produce higher specific/nonspecific (3H)-ouabain binding ratios, the (3H)-ouabain binding study, which is merely an estimation of active enzyme concentration, was performed at 37°C instead of 30°C. The binding reaction was started by adding the homogenate to the prewarmed incubation mixture and was terminated by filtering aliquots through Millipore filters which separated bound and unbound (3H)- ouabain. The filters were washed twice with 4 ml each of ice cold solution containing non-radiolabelled ouabain (0.2 mM), Tris-HCl buffer (pH 7.5, 50 mM) and 10 mM KC1. Millipore filters were then dissolved in ethylene glycol monomethyl ether and the radioactivity trapped on the filter (bound ouabain) was assayed using liquid scintilla- tion counting method. The liquid scintillation counting cocktail (4 g 2,5-diphenyloxazole (PPO) + 167 mg 1,4-bis[2-(4-methy1-5- phenyloxazoly1)]-benzene (POPOP) + 250 m1 ethylene glycol monomethyl ether (“Piersolve", Pierce) + 750 ml toluene) was used in the present study. Counting efficiency (approximately 30%) was monitored by the external standard channel ratio which was occasionally calibrated with internal standards. It has been demonstrated previously that ATP- dependent (3H)-ouabain binding is an accurate estimate of 25 Na+,K+-ATPase activity in the presence or absence of cardiac gly- cosides (Allen et_al,, 1971). Therefore, values obtained in the absence of ATP were subtracted from the total binding obtained in the presence of ATP to calculate the specific, ATP—dependent (3H)-ouabain binding. In one series of experiments, the (3H)-ouabain-enzyme complex was prepared by incubating the enzyme preparation or homogenate (0.2 mg of protein per milliliter) and 0.05 uM (3H)-ouabain in the presence of 100 mM NaCl, 5 mM MgCl 50 mM Tris-HCl buffer (pH 7.5) with or 2. without 5 mM Tris—ATP at 37°C for 15 minutes. The reaction was ter- minated either by the addition of non-radiolabelled ouabain (final concentration: 0.1 mM) with a simultaneous 5 or lO-fold dilution of mixture or by the addition of non-radiolabelled ouabain and a centri- fugation at 100,000 x g for 30 min at 0°C to separate unbound ouabain. The sediment was then suspended in 10 mM Tris-HCl buffer (pH_7.5) and the dissociation reaction was studied immediately. The dissociation of the (3H)-ouabain-enzyme complex was monitored either at 37°C or 27°C. The lower temperature would reduce the rate of ouabain release and thus allow for a more accurate measurement. Aliquots were taken at appropriate time intervals and the amount of bound (3H)-ouabain remaining undissociated was estimated using a Millipore filtration system to separate the unbound ouabain. Radio- activity was determined by liquid scintillation counting. The levels of ATP-independent binding, estimated by excluding ATP in the binding medium, were routinely subtracted from that observed in the presence 26 of ATP to calculate the ATP-dependent ouabain binding. All values re- ported are those due to the ATP-dependent portion of the binding. H. Sodium Pump Studies Sodium pump activity was estimated by means of ouabain-sensitive 86Rb-uptake according to the method of Bernstein and Israel (1970) with some modifications. Briefly, guinea pig ventricular slices (approximately 0.5 mm thick) were prepared using a Stadie-Riggs tissue slicer. After an incubation in Krebs-Henseleit solution with or without drug for various time intervals at 36:0.5°C, Slices were rinsed twice, transferred to a K+-free Krebs-Henseleit solution and incubated for 2 min at 0-2°C in order to "load" the slices with sodium so that intracellular sodium concentration would not be the limiting factor for the sodium pump activity. In one series of experiments, preincubation in cold media for "sodium loading" was omitted. Under these conditions, monovalent cation pump activity (the rate of ouabain- 86Rb-uptake) is influenced by intracellular sodium concen- sensitive trations (see Post, 1968; Thomas, 1972). The slices were then trans- ferred to a prewarmed, modified Krebs-Henseleit solution in which KC1 86 was replaced with 2 mM RbCl containing a tracer amount of RbCl. In 8°Rb, 0.3 mM ouabain was order to determine the nonspecific uptake of added to another vessel containing Slices prepared from the same heart. At appropriate time intervals, the slices were rinsed three times by immersion in a separate K+-free solution containing non- radiolabelled RbCl (2 mM) and blotted dry with filter papers. Radio- activity remaining in the slices were assayed using a Packard gamma 27 scintillation spectrometer. Values obtained from the slices con- taining added ouabain were subtracted from those obtained from slices without added ouabain to calculate the ouabain-sensitive 86Rb uptake, which is due to active cation pump activity. The ouabain-sensitive 86Rb-uptake was approximately 40-60% of the total uptake. Slices preincubated without the drug or with the vehicle served as controls. I. Miscellaneous Methods Protein concentration was determined either by the biuret method or that of Lowry gt_al, (1951). Bovine serum albumin was used as the protein standard. The time required for the development of 50% of the maximum positive inotropic effect were determined by probit analysis. Dissociation rate constants for the washout of the positive inotropic effects were calculated from a linear regression line fitted to the semi-logarithmic plot by the least squares method. J. Statistical Analysis Statistical evaluations of the data were performed by Student's t-test. Criterion for Significance was a probability of less than 5%. RESULTS A. The relationships amonggNa+,K+-ATPase, sodium pump activity and the positive inotropic response to cardiac glycosides Recently, considerable evidence has accumulated to suggest that the mechanism of the positive inotropic action of cardiac glycosides is related to the inhibition of cardiac Na+,K+-ATPase. Nevertheless, a causal relationship between these two phenomena is still not established because several investigators have reported that the inotropic responses of cardiac glycosides do not follow the time course of the inhibition of Na+,K+-ATPase during drug washout (Okita et 21., 1973; Ten Eick et a1,, 1973; Murthy §t_al,, 1974; Peters et_al,, 1974). Therefore, the relationships among cardiac Na+,K+-ATPase inhibition, sodium pump activity and the inotropic responses of cardiac glycosides were further studied using Langendorff preparations of guinea pig hearts to deter- mine if a causal relationship exists among these phenomena, particularly during the time course of development and diSsipation of the inotropic response. ‘ Isolated Heart Studies; Perfusion of 1.2 uM ouabain or 0.4 uM digitoxin for 20 min in the paced Langendorff preparations of guinea pig hearts produced 35:6 and 32:3% increases (mean i S.E.M. of 6 experiments) in isometric contractile force, respectively. When the perfusing solution was changed to drug-free solution, the positive inotropic response dissipated gradually. Fig. 1 shows the time course 28 29 Fig. l. The onset and duration of the positive inotropic action of ouabain and digitoxin in electricallyedriven (1-5 H2) isolated guinea- pig hearts. After a 45—min equilibration period, either 1.2 uM ouabain or 0.4 uM digitoxin was perfused for 10 or 20 min; 20-min drug-perfusion was followed by perfusion with drug-free solution for 8, 24 or 40 min, respectively. Each point represents the mean of at least 5 experiments. Vertical lines indicate standard error of the mean. Arrows indicate the time points at which perfusion was terminated for the experiments shown in the subsequent figures. 30 .3355. 05:. F k e h on 0* 00 ON 0— s s s s \ \-\-\- \- \-\g'\'\ '\-\-\-\ '\'\'\ d u q 1 \‘\‘\‘\$‘\!‘ 32:00 592553.}: F g 2 8 (woo no) Monoul «mood «so; .20 €22.20.— 930 i0... 2 1g. .228 .o + s... 3 o 31 of development and dissipation of the positive inotropic response produced by ouabain and digitoxin. Decay of inotr0py with ouabain was more rapid than that seen with digitoxin under these experimental conditions, but is consistent with earlier studies (Fricke and Klaus, 1971a; Akera gt_al,, 1973). Contractile force of non-treated hearts was remarkably constant for the total time of the experiment. At times indicated by the arrows, during both the development and dissi- pation of the inotropic response, hearts were removed from the appa- ratus and the degree of Na+,K+-ATPase inhibition was studied imme- diately. (3H)-0uabain Binding Studies. It has been demonstrated that, in guinea pig hearts, bound ouabain dissociates rapidly from the enzyme, even at low temperatures (Akera gt_al,, 1973). Thus, the degree of enzyme inhibition cannot be estimated by assaying Na+,K+-ATPase activity directly after the isolation and purification of the guinea- pig heart enzyme in contrast to earlier studies performed in the dog (Akera gt_al,, 1969, 1970). In order to study the correlation between the cardiac Na+,K+-ATPase inhibition and the inotropic action of the cardiac glycosides, as well as to study the nature of the cardiac glycoside-Na+,K+-ATPase interaction occurring in contracting isolated guinea-pig heart, the initial velocity of ATP-dependent (3H)-ouabain binding by homogenates was used as a means of estimating Na+,K+- ATPase activity (Allen e_t_a_l_., 1971b). As shown in Fig. 2, (3H)- ouabain binding in the presence of ATP increased with time, while the nonspecific ATP-independent (3H)-ouabain binding (approximately 0.05 32 400 '— ATP 5mM .300 .200 Bound (3H) Ouabain pmoles/mg proiein .100 Time (minutes) Fig. 2. The effect of ATP on the binding of (3H)-ouabain to homogenates obtained from isolated perfused guinea-pig hearts. Bound ( H)-ouabain is expressed as picomoles per milligram of protein. The incubation medium contained 5.0 mM MgC12, 200 mM NaCl, 0.01 uM (3H)-ouabain and 50 mM Tris—HCl buffer (pH 7.5), with or without 5 mM Tris-ATP. The reaction was started by the addition of heart homogenate (approximately 0.4 mg of protein per milliliter) to a prewarmed incubation mixture containing (3H)-ouabain. Each point represents the mean of 5 experi- ments. Vertical lines indicate standard error of the mean. 33 pmole/mg protein) was maximal after a 1-min incubation period and essentially unchanged at incubation times greater than that. At the end of a lO-min incubation period (3H)-ouabain binding in the presence of ATP was almost six-fold that of the binding in the absence of ATP. ATP-dependent binding, which is the difference between that observed in the presence and absence of ATP, was almost linear during the first 3 min (open circles, Fig. 3). Homogenates obtained from ouabain-treated hearts had a signi- ficantly reduced initial (3H)-ouabain binding velocity compared to that of control heart homogenates when the data were analyzed after 1 and 3 min incubations (closed circles, Fig. 3). However, the levels of (3H)-ouabain binding after a 10-min incubation were similar in homogenates obtained from both control and ouabain-perfused hearts. In addition, if the homogenates from glycoside-treated hearts were incubated in ouabain-free solution at 37°C for 4 min prior to the addition of the labelled ouabain, the initial (3H)-ouabain binding velocity of the homogenates obtained from ouabain-perfused hearts was similar to that of control homogenates (closed triangles, Fig. 3). This would indicate that the 4-min period was sufficient to allow the ouabain to dissociate so that the binding sites were now available for radiolabelling. Since the ATP-dependent (3H)-ouabain binding was linear during the first 3 min of incubation indicating that the value obtained at this time may be used as an estimate of the initial binding velocity (Fig. 3), this period for (3H)-ouabain binding was chosen in subse- quent studies. 34 ’Eé (3J3 F- .2 I E . i 5 ., (Lil - .0 E n- ‘E ii '5 JD 3 f, (IJ -— 9 E Q, [at 13 .. 1 c HE .c, l 1 l L .1 O 2 4 6 3 10 Time (minutes) Fig. 3. The ATP-dependent (3H)-ouabain binding of homogenates obtained from control and ouabain-perfused guinea pig hearts. The amount of bound (3H)-ouabain is expressed as picomoles of ouabain per milligram of protein. Values obtained in the absence of ATP (non-specific binding) were subtracted from those obtained in the presence of ATP (total binding) to calculate the ATP-dependent (3H)-ouabain binding. Open circles and closed circles represent the ATP-dependent ( H)- ouabain binding by the homogenate obtained from control and ouabain- perfused hearts, respectively. Closed triangles represent the (3H)- ouabain binding of the homogenate obtained from ouabain-perfused hearts and incubated in_vitro for 4 min in ouabain-free solution at 37°C before addition to the binding mixture. Vertical lines indicate standard error of the mean. Each point represents the mean of 5 experiments. *Significantly different from control; P<0.05. 35 The relationship between the ATP-dependent (3H)-ouabain binding of ventricular homogenates and the development and dissipation of the positive inotropic response during ouabain perfusion and washout is shown in Fig. 4 in which the initial velocity of the ATP-dependent (3H)-ouabain binding of homogenates obtained from control and ouabain perfused hearts and the corresponding percent change in contractile force was plotted against time in minutes. The (3H)-ouabain binding in control preparations was relatively constant which correlated well with its relatively stable contractile force during the course of the experimental period. An increase in contractile force following ouabain perfusion for 10- and 20-min was accompanied by a quantitative reduction of (3H)-ouabain binding velocity. Following a 20-min ouabain perfusion, the subsequent washout of inotropic response resulted in a recovery of (3H)-ouabain binding velocity toward the control values indicating that the ouabain which was bound to Na+,K+-ATPase during Langendorff perfusion dissociates rapidly from the enzyme during the period of drug washout in the contracting guinea-pig heart. The relationship between the initial velocity of the ATP-dependent (3H)- ouabain binding and percent change in contractile force was analyzed by the log-linear regression method and indicated a significant correlation (correlation coefficient; r=0.7l, n=34, P<0.01) between these two phenomena. If Na+,K+-ATPase inhibition is causally related to the positive inotropic effect of the cardiac glycosides, the same relationship should exist between Na+,K+-ATPase inhibition and changes in cardiac 36 1.2 1110411 000m 0mg Wash-out 140 T 0» ii £11: a. oi 52 m ii ca g 3 100 + W 5 I Control | cxuflwol E .200 i E 2 8 E .5 3 3 g .150 . ‘ “'1! :3 ° 8 I o. "’ " .100 n . 1 . 1 0 10 20 30 40 50 00 Time (minutes) F39. Effects of ouabain on cardiac contractile force and on initial H)-ouabain binding velocity. After a 45-min equilibration period, 1. 2 uM ouabain was perfused for 10 or 20 min; 20-min drug perfusion was followed by perfusion with drug-free solution for 8, 24, or 40 min. At times indicated by data points, hearts were removed from the apparatus and homogenized immediately. Homogneates were incubated at 37°C with 0. 01 pH (3 H)-ouabain in the presence of 200 mM NaCl, 5 mM MgC12, 50 mM Tris- HCl buffer (pH 7. 5) for 3 min with or without 5 mM Tris-ATP. ATP+dependent portion of ( H)-ouabain binding is plotted. Each point represents the mean of 5 experiments. Vertical line indi- cates standard error. 37 contractile force regardless of the differences in the time course of such events. Therefore, similar experiments to those described above were also repeated using digitoxin which has been shown to dissociate somewhat more slowly than ouabain from Na+,K+-ATPase (Yoda, 1973). In order to insure compatibility with the ouabain data, and since ouabain and digitoxin bind to the same site on Na+,K+-ATPase (Matsui and Schwartz, 1968), homogenates of digitoxin-treated hearts were challenged with (3H)-ouabain instead of isotopically-labelled digitoxin. Homo- genates obtained from 10-min digitoxin—perfused hearts also had reduced initial rates of (3H)-ouabain binding. AS with ouabain-perfused hearts, a 4-min incubation of the homogenates from digitoxin-perfused hearts prior to the addition of the (3H)-ouabain resulted in a partial recovery of the initial velocity of (3H)-ouabain binding (data not shown). Similar to the ouabain data, an increase in cardiac contractile force following digitoxin perfusion resulted in a decrease in the initial velocity of (3H)-ouabain binding in homogenates obtained from drug-perfused hearts (Fig. 5). And after a 20-min digitoxin perfusion, the subsequent drug washout resulted in a time-dependent decrease in contractile force and an increase in the initial velocity of (3H)- (ouabain binding; both returning toward control levels. A significant correlation (correlation coefficient; r=0.68, n=36, P<0.01) was also observed between the percent change in contractile force and ATP- dependent (3H)-ouabain binding. As shown in Fig. 1, the development and dissipation of the positive inotropic response produced by digitoxin 38 4x10'7 M Digitoxin Drug Wash-out 140T‘T ' ':Ti f 8 F 8 a s ': 8 E 120 o- 0 C I 1* :: " c: r o A: a; 10m) st—dl - Control Control 13 200 2 '5 g 9. o. ‘= :5 150» i '- r z E g .2 ° 2 :l: a. L “i " ‘umo ‘ ‘ ‘* L 0 10 20 30 40 50 60 Time (minutes) Fig. 5. Effects of digitoxin on cardiac contractile force and on initial (3H)-ouabain binding velocity. See legend to Fig. 4. The isolated hearts were perfused with digitoxin instead of ouabain. Each point represents the mean of at least 5 experiments. Vertical lines indicate standard error. 39 perfusion and washout were slower than with ouabain. However, despite the differences in the time course of onset and offset of the ino- tropic responses produced by ouabain and digitoxin, the relationships between Na+,K+-ATPase inhibition and inotropic response were similar as indicated by a significant correlation between these two phenomena (Fig. 4 and 5). These results strongly suggest a direct relationship between the Na+,K+-ATPase inhibition and positive inotropic effect of cardiac glycosides. 86Rb-Uptake Studies. Prior to this study, it had been suggested by some investigators that the inhibition of Na+,K+-ATPase pgr_§g_may or may not be sufficient or necessary to produce a positive inotropic effect (Nakamaru and Schwartz, 1970; Besch and Schwartz, 1970). There- fore, it was important to determine sodium pump activity during the development and dissipation of the inotropic response to the cardiac glycosides. 86 Ouabain-sensitive uptake of Rb by ventricular slices was used as a means of estimating the level of cation pump activity (Bernstein 86 and Israel, 1970). The values of Rb uptake in the presence of ouabain (non-specific uptake) were subtracted from those obtained in 86 the absence of ouabain (total uptake) to calculate the Rb uptake associated with cation pump activity. In Fig. 6, the sodium pump activity estimated from the ouabain-sensitive 86 Rb uptake and the percent changes in contractile force for digitoxin-treated and control hearts were plotted against the time in minutes. An increase in inotropic response following digitoxin perfusion for 20-min resulted 4O “to"! Mutt. Dru Null-out e a? 4 “° ' 1 Fhwwwunt13hmuql| In Contractile Foroo ‘ 8 100 Control I 3 40!! P ‘ ' 0 m1 3:3... °" 1 a P 3 " 1 . .2 52:! . 20K) - 2 8 3' m l l l L l (D 101 201 1301 '00) .501 UMP Tllno (inlnutoo) Fig. 6. Effects of digitoxin on cardiac contractile force and on ouabain-sensitive 86Rb uptake. See legend to Fig. 4. At times indi- cated by datum points, hearts were removed from the apparatus. Ventri- cular slices (0.5 mm thick) wags obtained using Stadie-Riggs tissue slicer, and ouabain-sensitive Rb uptake to estimate Na—pump activity was assayed immediately. After a 5-min preincubation of medium at 37°C for temperature equilibrium, the reaction was started by the addition of two ventricular slices whicg6 had been previously incubated at 0°C in K+-free solution. Munt of 6Rb uptake in the presence of 0.2 mM ouabain after 8 min were subtracted from that oggained in the absence of ouabain to calculate the ouabain-sensitive Rb uptake. gartical lines indicate standard errors of the mean of the changes in Rb uptake and changes in cardiac contractile force, respectively. Each point represents the mean of 5 experiments. 41 in a reduction of ouabain-sensitive 86Rb uptake. After a 20-min digitoxin perfusion, the subsequent washout of inotropic response also 86Rb uptake towards resulted in a recovery of the ouabain-sensitive the control values. Similarly, a significant correlation (correlation coefficient; r=0.70, n=32, P<0.01) was also observed between sodium pump inhibition and the positive inotropic response to digitoxin. Studies were also attempted with ouabain-treated hearts. Although pump activity was reduced, the data were not significantly different from control. This was undoubtedly because of the longer incubation time (8-min) required for the sodium pump studies, a time which is more than sufficient to allow ouabain to dissociate from its binding site in this species (Akera gt_al,, 1973, see also Fig. 3). Thus, it appears that the positive inotropic response of cardiac glycosides is intimately related to the inhibition of cardiac Na+,K+- ATPase activity. Furthermore, this inhibition of cardiac Na+,K+- ATPase actually results in a significant inhibition of the sodium pump activity. The magnitude of inotropic response and that of sodium pump inhibition is quantitatively related. 8. The relationship between the ouabain-Na+,K+-ATPase complex formed in the functioning heart and formed in vitro in the presence of various ligands Recently, several investigators have failed to obtain a positive correlation between the inhibition of cardiac Na+,K+-ATPase and the positive inotropic response to cardiac glycosides in the isolated, perfused heart preparations (Okita gt_al,, 1973; Ten Eick gt al,, 42 1973; Peters gt_al,, 1974). According to these investigators, the enzyme inhibition persisted whereas the positive inotropic effect dissipated rapidly. A number of other investigators, however, have reported that ouabain, strophanthidin-3-bromoacetate and strophan- thidin are all reversible inhibitors of the Na+,K+—ATPase jn_yjtrg_ (Akera and Brody, 1971; Fricke and Klaus, 1971b; Akera gt_al,, 1973; Tobin et_al,, 1973). Thus, there was a possibility that the cardiac glycosides-ATPase interaction in the contracting heart may be different.from that occurring jn_vjtrg_with partially purified enzyme preparations. Therefore, the relationship between the ouabain-ATPase complex formed in the beating heart and those formed in_yjtrg_in the presence of various ligands was studied. Cardiac glycosides, such as ouabain, preferentially bind to Na+,K+-ATPase in the presence of a variety of combinations of ligands and the resulting ouabain-enzyme complexes formed jn_yjtrg_seem to have different properties depending upon the ligands present during the binding reaction. The differences in characteristics of the ouabain-enzyme complexes may be detected by monitoring dissociation rates in media of low ionic strength (Akera and Brody, 1971). The, it is possible to compare the characteristics of the ouabain-ATPase complex f0rmed in the contracting heart and that formed jg_yjt§9_with partially purified enzyme preparations by monitoring the dissociation rate constants. Figure 7 shows the rates of the release of (3H)-ouabain pre- viously bound to a dog heart homogenate at 37°C under various con- 43 Fig . Dissocia 10" of( 3H)-ouabain from dog heart homogenate. Dog heart7 homogenate (O. 2 mg of protein per milliliter) was incubated with 0.01 uM (3H)- ouabain at 37° C for 20 min in the presence of either 20 TM NaCl, 5 11M MgClz and 5 mM Tris—ATP (O, O), or 20 nil NaCl, 5 mM KC1, 5 mM MgC12 and 5 mM Tris-ATP (EJ,II). The mixture was centrifuged at 100, 000 x g for 30 min at 0°C, resuspended in 10 mM Tris- HCl buffer (pH 7. 5) and assayed for the release of (3 H)-ouabain at 37° C in the absence (open symbols) or presence of 5 mM KC13(filled symbols). Values are expressed as percentages of bound ( H)-ouabain at the start of the dissociation reaction. Vertical line indicates the standard error of four experiments. 44 100 90 70 60 at O .5 O l / 30r- 20- Bound I 3lll-ouabain ( por cont) :1 I Na+,K+,M92+,ATP oo Na+,M92+, ATP r-O'I ‘0 I l 0 10 20 30 Time (minutes) 45 ditions and dissociated at the same temperature in 10 mM Tris-HCl buffer (pH 7.5) in the presence and absence of K+. In these experi- ments, (3H)-ouabain was incubated with a dog heart homogenate in the presence of Na+, Mg++ and ATP or Na+, K+, Mg++ and ATP for 20 min at 37°C. Subsequently, unbound (3H)—ouabain was removed by centrifuga- tion. Following the wash and resuspension of the pellet and subsequent dissociation of the bound ouabain was monitored at 37°C in 10 mM Tris- HCl buffer (pH 7.5). The dissociation half-time of the (3H)-ouabain- enzyme complexes formed in the presence of Na+, K+, Mg2+ and ATP was 2* and ATP (Fig. 7). slower than that formed in the presence of Na+, Mg However, the addition of 5 mM KC1 stabilized both complexes to the same degree. The (3H)-ouabain-enzyme complex, prepared from a partially puri- fied dog heart enzyme in the presence of Mg2+ and Pi, however, showed a slow dissociation at 37°C in the absence of K+ and the addition of 5 mM KC1 to the dissociation mixture failed to Significantly in- fluence the stability (data not shown). Thus, it appears that the ouabain-enzyme complex formed in the presence of Mg2+ and Pi has different characteristics from that prepared in the presence of either 2 2+ and ATP which was K+-Sensitive and Na+, Mg + and ATP, or Na+, KT, Mg was maximally stabilized by this cation. In order to study the relationship between the ouabain-ATPase complex formed jg_vitro with partially purified enzyme preparations and those formed in the beating heart, similar studies were performed following gx_vivo treatment with radiolabelled ouabain. In isolated 46 puppy heart Langendorff preparations, a perfusion of 0.6 uM (3H)- ouabain for 20 min produced a 29.1:l.9 percent increase in isometric contractile force (mean : standard error of four experiments) when the heart was perfused at 30°C and electrically driven at 1.5 Hz with a resting tension of 4 g. At this time, the ventricular muscle was excised, homogenized, centrifuged at 100,000 x g for 30 min at 0°C, the pellet resuspended in 10 mM Tris-HCl buffer (pH 7.5) and assayed for the release of (3H)-ouabain at 37°C. The release of (3H)-ouabain in the absence of added KC1 was relatively slow; the half-time was approximately 25 min (Fig. 8). It should be pointed out that values in Fig. 8 represent "total" ouabain binding to the particulate fraction of ventricular tissue. Although it was not possible to distinguish that portion of (3H)-ouabain which was specifically bound to Na+,K+- ATPase from that non-Specifically bound to the tissue under these experimental conditions, the addition of 5 mM KC1 markedly reduced the rate of the release of (3H)-ouabain. The effects of 5 mM NaCl to reduce the rate of the release of (3H)-ouabain were significantly smaller (Fig. 8). These differential effects of Na+ and K+ on the release of bound ouabain closely resemble those on the dissociation of the ouabain-Na+,K+-ATPase complex preformed jg_vjt§9_in the presence of Na+, Mg2+ and ATP (Akera and Brody, 1971). While possibilities exist that K+ non-specifically alters the rate of release of bound ouabain which is not associated with Na+,K+-ATPase, the differential effects of K+ and Na+ observed strongly suggest that the K+-sensitive portion of bound ouabain is associated with Na+,K+-ATPase. The ATP- dependent portion of bound (3H)-ouabainformed in the presence of Na+ 47 60v- 1 100 T Y I r I '— 3; - KCl 3 80 - 5m" 1 o 3 o I: '3 7O - - f; 3 NaCl )— 5mM I ‘1’. 'o I: :3 o In Control 40 A 4 J L 1 L 0 5 1O 15 20 25 30 Time (minutes) Fig. 8. Dissociation of (3H)-ouabain from ventricular tissue after puppy heart perfusion of the drug. Isolated puppy hearts were perfused with 0.6 uM (3H)-ouabain for 20 min at 30°C. After perfusion of the drug into puppy hearts, ventricular muscle was homogenized, centrifuged at 100,000 x g for 30 min at 0°, resuspended in 10 mM Tris-HCl buffer (pH 7.5) and assayed for the release of (3H)-ouabain at 37°C in the absence or presence of either KC1 or NaCl. Values are expressed as percentages of bound (3H)-ouabain at the start of the dissociation reaction. Vertical line indicates the standard error of eight (control and KCl) or four (NaCl) experiments. 48 and M92+ has been shown to be associated with Na+,K+-ATPase (Allen gt al., 1971), whereas the ATP-independent portion is considered to be nonspecific. While the dissociation of the ATP-dependent portion of bound ouabain is differentially affected by Na+ and K+ (Akera and Brody, 1971), that of the ATP-independent portion is affected neither by Na+ nor K+ (data not shown). Furthermore, the release of bound ouabain during the jg_vjt:9_incubation was accompanied by recovery of the Na+,K+-ATPase activity estimated from the initial velocity of (3H)-ouabain binding (data not shown). Therefore, it appears that the (3H)-ouabain bound to Na+,K+- ATPase during the perfusion of isolated heart and the complexes formed 2+ jg_vitro with Na+, Mg and ATP in the presence and absence of K+ had similar characteristics and were both reversible and stabilized by K+. C. The relationship between sodium pump inhibition and myocardial contractile force If Na+,K+-ATPase inhibition is causally related to the positive inotropic effects of cardiac glycosides, the same relationship should- also exist between ATPase inhibition and changes in cardiac contrac- tile force, regardless of the mechanism of enzyme inhibition. Na+,K+- ATPase inhibitors such as N-ethylmaleimide (Skou, 1963; Skou and Hilberg, l965), p-chloromercuribenzoate (PCMB) (Skou and Hilberg, l965), fluoride (Opit gt_al,, 1966) and low temperature (Repke, 1964) have been shown to produce positive inotropic effects (Bennett _t_a1,, 1958; Yamamoto, 1967; From and Probstfield, 1971; Covin and Berman, 1959; Berman, 1968). These agents or conditions, however, produce 49 other biochemical effects as well, and it is difficult to determine if the observed changes in contractile force result from cardiac Na+,K+- ATPase inhibition. Na+,K+-ATPase is an enzyme which is activated by the simultaneous presence of Na+ and K+. Several monovalent cations can either substi- tute for or compete with Na+ and K+-activation of the enzyme (Skou, 1960). Recently, it was demonstrated that Rb+ inhibits Na+,K+- ATPase in the presence of Na+ and K+ (Tobin et_al,, 1974). Since several monovalent cations with differential effects on Na+,K+-ATPase activity are available, the effects of monovalent cations on cardiac Na+,K+-ATPase activity and cardiac contractile force were investigated. Na+,K+-ATPase Activity. The effects of monovalent cations on cardiac Na+,K+-ATPase activity are shown in Table 1. In this series of experiments, the Na+,K+-ATPase activity of partially purified enzyme preparations obtained from guinea pig hearts was assayed in the presence of 150 mM Na+ and 5.8 mM K+, an approximation of extracellular cation concentrations. Control Na+,K+-ATPase and MgZ+-ATPase acti- vities were approximately 15 and 10 umoles Pi released from ATP per milligram protein per hour, respectively. The addition of 3-10 mM Na+ produced a small but statistically significant (P<0.05) stimulation of enzyme activity, while the addition of 2-5 mM K+ or 3-10 mM Li+ failed to significantly alter Na+,K+-ATPase activity under these experimental conditions. Addition of Rb+ caused a significant inhi- bition of the enzyme activity at 3 and 5 mM. Tl+ was inhibitory above 1 mM but not at lower concentrations. The addition of Cs+ and NH4+ produced little or no effect on the enzyme activity. 50 .35.: V k v .828 So... acouoté 323...:Em m ._.\=.3o.:. “:55 8.9:: mg 98>» 85.38: .9555 $329,895.. 53.3.22. .5: :c..§._::.am 38.65 29.82: o>3awo= was ozfimom .wacofitomxm m. .«C ...—fawn .2: MM _O.—a-OO GA: EOHM mefida—O ”tawny-0AM mew fiOmehAHNO th w®~£d> :4 a maize :8 3 Es. aw: :6 3. pm as .3. .--:V. 7.5 :3 ..o 09:89:. 23 :. womamma 33 5.2a?» 02584-3. .+aZ a «Hm an: ””3: an: .57. mum. macho 3:. so #13:... $th «he! Emmi ..-a. mafia... alum... «Pal LE ..nm .Hml «H»... t: .Hm IS an: .2 Amha— éHm— mwuumfi gaunt ...wVi 2:. ...... 7.5 :6 SE ...” SE QN 28 c4 28 ad 25 and 1.233: awash}? . v— .+az 06.6.30 :0 £8.38 e:o.d>o=oE mo Boohm A finch. 51 Since the inhibitory effects of monovalent cations on ATPase may be due to a competitive displacement of Na+ ions (Skou, 1960), a greater inhibition may be observed in an assay medium containing a lower concentration of Na+. Moreover, in intact cells, Na+,K+-ATPase is activated by intracellular Na+ and extracellular K+ (Glynn, l962; Whittam, l962; Baker, l963). Thus, a medium containing 20 mM Na+ and 5.8 mM K+ may be more relevant to isolated heart studies than that containing l50 mM Na+ and 5.8 mM K+. Table 2 shows the effects of monovalent cations on Na+,K+-ATPase activity assayed in media con- taining 20 mM Na+ and 5.8 mM K+. Control Na+,K+-ATPase activity under these experimental conditions was approximately 10 umoles Pi/mg protein/h. The addition of 3-10 mM Na+ and 3-l0 mM Li+ produced a slight but statistically significant stimulation. The addition of Rb+ and Tl+ under these conditions again produced a significant inhibition of the enzyme activity. K+, Cs+ and NH4+ failed to produce any marked effect on enzyme activity. It has been observed that the inhibitory effect of Tl+ on the ATPase activity of red cell membranes was more pronounced in fresh, low Na+, high K+ red cells than in cold-stored, high Na+, low K+ cells (Skulskii gt_al., 1973). When guinea pig cardiac Na+,K+-ATPase acti- vity was assayed in the presence of 75 mM K+ and 20 mM Na+, an approxi- mation of intracellular cation concentrations in cardiac muscle, the control Na+,K+-ATPase activity was lower than that in low potassium media and approximately 2 umoles Pi/mg protein/h. Under these assay conditions, Tlf exerted a dose-dependent inhibition of ATPase activity (Table 3). The inhibitory effect of Tl+ was significantly less in the ..mod V at .9555 59... 5555...... 3555...:wi .r. 5.5.39... w5\.m 5.5:: o“ 953 5.5.3555 .9555 5555.55.95 m. ..c ..m.m.m 2.. mm .9555 2.5 59... 5585.5 5555.. 5 «53935 5.5 53...; :4 s .93. .5. 25 an as. am: 25 5.. .2 7.... .5... .... :7. 25 om .0 555on 2.5 :. 69mg 33 5.35.5 83fl8<¥¥ $57... a in. sum: in» in” ...my. “Hm 33.: who .8 a wing! anvil .3157 131 +5. final .13.... mammal LE .mufi af: 3%.. .3 3...: Ida an: -m ..xnfi an”. .3”: an; .5. 7.... ...:— 7.E ed 25 ad 25 ed 25 c4 25 ad 2.5 mm... .79.». 55 85h8<¥¥ ..EZ 55.6.55 :0 58.55 50.55555 .0 35..."... .N 53:8 53 presence of l50 mM Na+, instead of 20 mM Na+, in the assay medium (Table 3). We can conclude from these data that among monovalent cations, only Rb+ and Tl+ consistently produced an inhibitory effect, whereas Na+ and Li+ stimulated the Na+,K+-ATPase activity in a low sodium assay medium. K+, Cs+ and NH4+ produced essentially no effect on ATPase activity. The inhibitory effect of Tl+ appears to be antagonized byNa+. Isolated Atria Studies. Contractile force of control prepara- tions was stable during the 30-min experimental period following a 60- min equilibration (data not shown). Cs+ failed to alter the isometric contractile force significantly at the concentrations of up to 10 mM (Fig. 9). The addition of NH4+ to the bathing medium, however, pro- duced a transient positive inotropic response lasting 2-3 min (Fig. 9). Addition of K+ also produced a transient positive inotropic response which was followed by a negative inotropic response (Fig. 10). The initial transient positive inotropic effects of K+ and NH4+ were partially blocked when l0 uM propranolol (final concentration) was added to the media 20 min prior to the addition of NH4Cl or KCl (data not shown). The addition of 2-5 mM Rb+ produced a dose-dependent increase in contractile force which developed rapidly, reaching a plateau in l-Z min (Fig. 10). Prior addition of l0 uM propranolol (final concentra- tion) did not significantly alter the sustained, Rb+-induced positive inotropic response (Fig. 10), indicating that the major portion of the inotropic response to Rb+ is not mediated by a beta-adrenergic mechanism. 54 .399 V am. 3.5.85 895 5:95:35 .m.5:aom.:w.m m .m 83:58 :. $5.393 wEFH 3.5813 9:. 4 5:55.: a. £539... wash 5.9:: N "953 85.355 .9555 55852.5 w .5 ..mdfl 2.5 ...... .9555 2.5 895 mamaano £595.. 3 585.558 98 £5.59 :4 a .meI mmHfiI 3. m H m I an H a I 3 a H a I a H 3 I 3 a H 5 I a H a I ....o 5 H a I a. H 5 I and a. .3 28 2:6 E ....z 25 a: .0. 28 2. .....z 282.. ..a. «.3352. enam9<.+vm .+dz 53.53 .3 no. 25:5.» .5 magnum .m 5369 55 Fig. 9. Effects of Cs+ and NH4+ on electrically dri¥en guinea pig+left atrial strips in Krebs-Henseleit solution (l45 mM Na and 5.8 mM K ) at 30°C. Atria were electrically stimulated with two platinum field- electrodes at l Hz. Following a 60-min equilibration period, one concentration of either Cs+ or NH + was added to the bathing media at 0 min. The changes in iSOmetric contractile force were expressed as percentages of the pre—drug value. Vertical lines indicate standard error of the mean. Each point represents the mean of at least 5 experiments. 56 ."'I ,.r \ A :50 I! II I \ \~ [I], I \ K ‘ I 11/ I H x 5 5 Ifi/ I .III H\ 5 5 I 5 zu/H .H\ 5 .. I 'l'l' \ \ I A 55 «.0 xx 5 :2 ll \ ... // x I Op. or O“ on 00103 omoenuoo u! caucuo wowed 57 Fig. l0. Effects of K+, Rb+, and Li+ on electrically driven guinea pig left atrial strips in Krebs-Henseleit solution at 30°C. Other conditions are same as in Fig. 9. ( ..... ) indicate the presence of 10 pH (:)-propranolol added 20 min prior to the addition of 5 mM Rb . Vertical lines indicate the standard error of the mean. Each point represents the mean of at least 5 experiments. 58 35m ZEN 25..” SEM 35m SEO— SEN SE” 28m 38m 00103 omoumuoo u; caucus wowed 59 It has been demonstrated that the development of the positive inotropic action with the cardiac glycosides is slow and is deter- mined by the number of contractions rather than by the time of ex- posure (Moran, l967). However, when similar experiments were per- formed with Rb+, no contraction-dependency for the positive inotropic effect of Rb+ was observed (Fig. ll). The inotropic response to Rb+ developed very rapidly. In Fig. llB, the percent maximal change in the inotropic response to 5 mM Rb+ at 30°C is plotted against the number of contractions. The numbers of contractions required to reach 50% of the maximal inotropic effect are significantly different for different stimulation frequencies. Moran (1967) has shown previously .that comparable curves for ouabain are superimposable when the ino- tropic response is plotted against the number of contractions, thus demonstrating that the development of the inotropic response to ouabain is dependent on the number of contractions, but not on expo- sure time. Clearly, such a beat dependency does not exist for the Rb+-induced inotropic response. The addition of 0.25-2.0 mM TI+ to the bathing medium produced a dose-dependent increase in the contractile force which developed gradually over a 30-min period (Fig. l2). The presence of 10 pM propranolol in the medium did not alter the positive inotropic response (data not shown). Differences in the time course and magnitude of the positive inotropic effects with Tl+ and Rb+ would suggest that the actions of Rb+ and Tl+ may not be identical. Supporting this, the onset of the inotropic response to Tl+ was significantly enhanced by 60 Fig. ll. The positive inotropic effect of Rb+ on isolated, electrically driven left atria of guinea pig heart as a function of time (A) and + the number of contractions (B) at 30°C. Conditions as in Fig. 9. Rb (5 mM) was added at 0 min. The changes in contractile force after the addition of Rb+ have been normalized on a basis of percent of maximal change and were plotted against time (A) or number of contractions (B). Each curve represents mean data from six atria. The difference between the curves in both plots are statistically significant at P<0.05. 61 28:92:58 .0 i .00.. 2...... 8- on up 0 00— on o— a «qqqfid q 5 O 5 8 8 8 mum mmnxuu so MIOOJ.‘ mono; emanation 62 Fig. l2. Effect of Tl+ on electrically driven guinea pig left atrial strips in Krebs-Henseleit solution at 30°C. Conditions as in Fig. 9.+ Following a 60-min equilibration period, a single concentration of Tl was added at 0 min. The changes in isometric contractile force were expressed as a percentage of the pre-drug value. Vertical lines indicate the standard error of the mean. Each point represents the mean of at least 5 experiments. 63 2.0m“ 1 0 mi 0.5 “J 0.25mN-I J l l 150 120 '- 8 8 8 09105 omocuuoa u! caucuo was 10‘ TimImIn) 65 reducing the frequency of stimulation from l Hz to 0.25 Hz (data not shown), in contrast to the onset of the inotropic response to Rb+, which was slightly inhibited by reducing the frequency of stimulation (Fig. llA). A possible explanation for this finding is that Tl+ may be acting at an intracellular site and the cell membrane may be permeable to Tl+ only during the resting state when the transmembrane electrical gradient is steep. It has been suggested that Tl+-transport into erythrocytes is a passive flux (Skulskii et_al,, 1973) and may be driven by chemical and electrical gradients. In order to further investigate this possibility, the following series of experiments were conducted. After the atrial strip was equilibrated for 60 min at l Hz, l mM Tl+ was added to the bathing media for a l5-min period during which the electrical stimulator was turned off (Fig. 13). The bath fluid was then rinsed twice, and the stimulator was turned on. In control experiments in which the stimulator was turned off without adding Tl+ to the bathing media, an initial large contraction was followed by contractions which were markedly reduced upon the reinsti- tution of the electrical stimulation compared with those observed before the quiescent period (Fig. l3, upper tracing). The magnitude of contractile force gradually increased during a 5-min period follow- ing the resinstitution of electrical stimulation, demonstrating a positive staircase phenomena, and at 5 min, the magnitude of contrac- tion was slightly larger (24.0:2.5%) than that of the pre-quiescent period but decremented to only l4.5i4.9% above the pretreatment level after 15 min. In experiments in which l mM Tl+ was added to the 66 Fig. l3. Effect of Tl+ on isolated, electrically driven left atria of guinea pig heart when exposure to the drug was restricted to non- contracting tissue. Following a 60-min equilibration period at l Hz, Tl+ (final concentration, l mM) was added to the bathing media for 15 min during which the electrical stimulator was turned off (lower tracing). The bath was then rinsed twice, and the stimulator was turned on. In control experiments, the stimulator was turned off without adding Tl+ to bathing media (upper tracing). The contractile force of Tl+ treated and washed atria was significantly higher than that of control following resinstitution of electrical stimulation. A typical tracing from 5 experiments in each group. 67 :0 ...-=0 :0 .3530 0.0 0.. (“I can: 68 bathing media during the l5-min quiescent period, a similar pattern of post-quiescent contraction was observed. The magnitude of contraction, however, was significantly greater in Tl+ treated and washed atria during the post-quiescent period. At 5 min after the reinstitution of electrical stimulation, the contractile force of Tl+ treated and washed atria was 66.l:8.4% above the pre-quiescent level. The positive inotropic response did not significantly decrease after 5 min despite the absence of Tl+ in the bathing media and was maintained at 64.7i7.7% above the pretreatment level at l5 min. These values are significantly different from those of respective controls at P<0.0l. These data indicate that the effect of Tl+ is not easily reversible. If a causal relationship exists between sodium pump inhibition and the positive inotropic effects of Tl+, Rb+ and cardiac glycosides, one would expect a quantitative correlation to exist between these two phenomena. However, the extremely slow washout of the Tl+-induced inotropic effect (Fig. 13) tends to support the hypothesis that an intracellular accumulation of Tl+ is required for its action. Such an event cannot occur in partially purified enzyme preparations since these preparations do not possess limiting membranes and Tl+ cannot accumulate. Consequently, jn_vitro studies would result in a marked underestimation of Na+,K+-ATPase inhibition by Tl+ which would occur in beating hearts. Thus, the effect of Tl+ on sodium pump activity was estimated by determining the ouabain-sensitive 86Rb uptake of guinea pig ventricular slices. Preincubation of ventricular slices with l-4 mM thallous acetate resulted in a concentration and time 69 86 dependent reduction in the ouabain-sensitive Rb uptake compared with their respective controls (Table 4). Ouabain-insensitive 86Rb uptake, i.e., 86Rb uptake observed in the presence of 0.3 mM ouabain, was only minimally affected by Tl+ pretreatment. It should be noted that ventricular slices were exposed to Tl+ only during the preincubation period and the subsequent 86Rb uptake studies were performed in the absence of Tl+; the slices were rinsed three times with Tl+-free Krebs—Henseleit solution prior to uptake studies. Intracellular Tl+ concentration in these slices, determined by atomic absorption, indicated that slices accumulated Tl+ in a concentration and time dependent manner (Table 4). g The addition of Li+ to the bathing media produced a small, but significant positive inotropic response at l0 mM but not at lower concentrations (Fig. l0). Additional studies with Li+ at higher concentrations (30 mM) showed a gradual increase in contractile force reaching 38.5:3.4% above control amplitude after 30-min exposure (Fig. l4). The presence of 10 pH propranolol did not significantly alter the positive inotropic response. The addition of an equimolar amount of sucrose (60 mM), however, produced an immediate increase in con- tractile force reaching a plateau in 2-3 min (data not shown). Therefore, it is not clear whether the positive inotropic responses shown in Fig. l4 are due to the specific effect of Li+ or are merely osmotic effects. However, differences in the pattern of onset of inotropic responses to these two agents appear to suggest that the response to Li+ is not the result of an osmotic effect alone. 70 ..mc.ovmv pogucou scum «cusuwwwu Apucuupmvcmwmc .:.m.m H and: .muco555oaxo mo gangs: 0:» monou5uc5 monocucmsoa :5 Lucas: .mco55osucoucou +5» c.5355ouoguc5 so oxouaaiaxom so» gosa5o nomamma cog» use gowns—om «.mpomcozimaogx outwi+x s55: uo«:.g..+.h mo mucous. so oucomugn use :5 comm 5a covenauc5ogn «so: mou55m gopau5gacu> 53 55.23.. .3 $.38... .3 5.2%.” +55 55... e E 55.2.3.5 E $5535.. +55 5.. .5 .8 55.55.53 58 2.38.5 .8 55.9.8.5 +55 5.. N .3 55.385 .55 55.9.8.5 +55 .5... N :55 2.35.5 .8 555.2255 53 «5.555555 +55 55. 5 .3 8.33.5 S $5.28; +55 5... 5 a: 2.92.... a: 2.28.... 58:3 9- ..5. 8 5.. 55 e5... 5 Essay»... 5... 8 5.5.. 3 Essay»... 5 . ...5755533: 235. 852.... .o : pxcus. +5» 5.5:..ouoguc. oxaunawmmmmwwwu5mcomuc5oaoac ‘ mason; o_a soc—aw =5 co5uacuguucou co. mac—pug» 5.5355ouoguc5 we. oxuuaainuoo o>5u5mconic5~acao co «co. mac—5agu yo uuoeuu .. u4m " 1.5 J! lg 1.0 0.5 0'0 I l l 71,] l 1 l Propranolol 0 5 ‘0 2° 3° 10 ”M ( Time (min) Sucrose 40mM 2.0 1.5 3 § 1.0 o “' 0.5 0.0 L 1 EFL—51; 5 10 20 30 Propranolol 10 pM Time (min) LI" 2()nMH Fig. 15. Typical tracing of the effects of sucrose— and Li+-substituted Krebs-Henseleit solution on isolated, electrically driven guinea pig left atrial strips. Conditions as in Fig. 9. Following a 60-min equilibration period and a 20-min pretreatment with 10 pH (:)-propranolol, 20 mM LiCl was substituted for molar equivalents of NaCl in the regular Krebs-Henseleit solution (145 mM Na+ and 5.8 mM K+) at 0 min (lower tracing). 40 mM sucrose substitution for 20 mM+NaCl served as control (upper tracing). After 30-min exposure, the Li —substituted solution produced a significantly greater inotropic response (52.l:4.6%) than that of control (23.2tl.9%; mean :SEM of five experiments). 75 It has been reported that Li+, when substituted for NaCl, enters the intracellular space and accumulates intracellularly because the sodium pump cannot remove it effectively (Keynes and Swan, l959) resulting in an inhibition of sodium pump activity (Bose and Innes, l973). Thus, the effect of Li+ on 86Rb uptake, an estimate of sodium pump activity, of guinea pig ventricular slices were conducted. Preincubation of ventricular slices with 29-87 mM LiCl (equimolar replacement of NaCl in the Krebs-Henseleit solution) resulted in a concentration and time dependent reduction in the ouabain-sensitive 86Rb uptake compared with their respective controls (Table 5). 86 Ouabain-insensitive Rb uptake was also slightly depressed by Li+ pretreatment. It should be noted that ventricular slices were exposed to Li+ only during the preincubation period and the subsequent uptake studies were performed in the absence of Li+. In addition, the slices were also preincubated for 2 min at 0°C in order to "load" the slices with Na+ so that intracellular Na+ concentration (approximately 90-l00 mEq/l. H20) would not limit the pump activity. Intracellular Li+ concentration in these slices, determined by atomic absorption, indicated that slices accumulated Li+ in a concentration and time dependent manner (Table 5). Therefore, it appears that among monovalent cations, only sodium pump inhibitors produced a sustained positive ino- tropic response. D. Comparison ofgthg_effect of monovalent cations and ouabain on cardiac Na',K'-ATPase and contractile force If the inhibition of cardiac Na+,K+-ATPase is the mechanism of the positive inotropic effect of Rb+, Tl+ and cardiac glycosides, 76 ..mo.ovav pogucou Eon» ucmsomwwu xpucoummwcmwma .z.m.m 5 com: .mucoewguaxm 50 Lopez: 6:» mmuoupucp m5mmgucmcaa :. cmnssz .copuupassuuo +55 cops—pouucuc5 so mxouaainmwm so» gogapm umaommm ems» van :o5535om 55m—wmcmzimnmcx o eac5-5 e55: come5c ..5555 .e e055ac5ed u tapas vmumu5uc. :55: Fonz 5o acmsouapaoc gmposwaco zuplv ccvua—Om HwflpflmC0zumn0o-x mewvg LO PMs—LO: .2. 000” an Bugaucpoga 0L0! mouwpm mezuwbucm> .55. 55.5.eo.~m .555 55.5.oo.m~ .55. 55.5555.e~ +55 :2 5m .05. 55.5.55.n~ 5°55 55.5.oa.m5 .c5. 5m.cn5~.55 +55 :5 mm .55 .o5.ohm~.5 .0. .55.ohmm.5 .55 .m5.ohm5.5 ..5 :5 5m .55 .o5.ofimm.~ +_5 :5 mm 5555 5e.5.m5.~5 .55. 55.ehme.o5 .55. 55.5.55.a +_5 as am .55 e5.onoc.~ .5. o~.odoe.~ ..5 ze 55 .555 m5.ehm~.~ .55. 55.0.oe.~ 5ee5eeu weph oe_h =_s cm :52 m. :53 m :o5aonauc5osa :55 on :55 m5 =055un=uc_oca .55: .553535 +55 c.5355euee5e5 5e5eim55d=mm55 52555 me5eee5 mxnuaainmm «>555mcom-:_unoao magma: a5; nocpaa :5 covuncucoucou +5. capappouocu=5 ecu oxuuaaiaxow «>555mcomic5unoao co +55 5o uuocwu _m u4mwuumammc .m so.oeco.o can >e.~.mho.emi .>e m.~nm.~o mew; Fopocmcaoea mo mucmmmga any cw :owmcmu uwgumeomw ucm meucmuoa m:_ummg .mu:u_pasm pmwucwuoa cowuum no» mwapm> Pogucou .pocucou we ”amoeba one we commmcaxm mam mmapm> .omucm mm: "H xhuizmie co H xhu cmzawm swam. cps om new E=_ume mg» a» vague mg: popocmgaogaip.u z: op .cowumcnwpwscm cmumwuum mmmapPuua mmmab .mewxoo gum: comm “a cowuansacmsta =_s-om a tspcwupmcom-:_mnm:o mace AmcmvmoPIE=muom uaogumzv mxmua: am m>wummcwmicwmnm=o co mcmxouocmamgm so uummmm mm m u4m .mcwxoo gum; goon on cowumo=u=+msa cws-om a tascwumm=mmicwmao=o mace Amcmumcpiszwuom gawzv mxmaa: am m>wupmcmmicwmnmzo co mcwxopccmawcm mo pummcm mm m m4m