i I -l I 2 1 .1 :2 . .3, ‘Qufi-E‘T-é?".t.f ‘ “‘j, _ . ( .cI‘: fl'Fuuh ”2323‘" ,’U( 321;: 3'»; $51.. ‘3 3 ‘31:??235‘ 52“: “-1 :52} II a‘éfia . M _ .— flax .-< , ‘2" . r "1 *. 5(3 2 335 A?“ '2 I‘ “Q3“ ,3, I. I? - 253‘“ 222 , 222M {:21 33.31332153323333.‘ 5333. I 51") 2 22 1‘2.2523533 323 “52", 3, ....(....;_ :5. 133*” ‘ 2 A .,. 722‘ - ~ ~ ' ‘ 51.. ‘Z‘ ' .24— 7. ... . - - ‘. ""ii , ,r __ . [4&3— :: . 1’3?“ ,9? ._ _. - N ‘ ,L ¢.. ' '. . ~ in. 2;; . « . . ; . ' 4.7.1:“ 4: .. .fi" .. .- .I ' 'J ‘ i2 22’ 5.31 522 '5 '1: I . . ‘ I" 3333‘2 II 2.23 222322? 3239,. 23$] £22m“ ‘7“! 922333. «22:22, 322222. 2222123 F” 17.3an 4:52- ,r 2’3 I I “w .:= '5; 229.33.55.15..;J '3. ’33:": $5?ng ”I” :1, "“L-i $5.95“ ‘ » . 41—5551,? +5233 ‘6? ‘ P. g... .512 .1 t, ‘1; ‘ I "‘43:- :.,V 1 :31. ;_§.;-r::l var: fi- 3:436:11. 15.145.“ 1 H :1 - 73 L 2.; ?§ ~ “'37": . F“ #3:?” Q5; .4: t 5.31425: 143933?” 3:, A" ‘ ' 41.91"; 5-..... ‘Im *3“..th ‘3 V :4._.4‘g ast’.‘ .' - ’. _2£:c~_.¢.. {33:57 “:1? "’2“? "‘ dean ... 1:. I?!” .I‘N fl" . . J- .2 =3» I, >355?" ‘ " - . .3“ . w 5"; ' - ' L“ .; *rihfi ._ .« _. .-.;‘Z‘~‘$‘?-‘ " ‘ ~ -, " $115.77" ‘ - _ £5733; . ‘ 3 .. ;- « . . ‘ ‘Er ‘ ’7‘.“ "‘ 133-5- if}? ' :5; - :l—‘YL’. __._..-.V.. , 4;- ~ vex ; ~2- 1? L “#4. 2 . ”v 1:1; . v—A‘I 514": - ix -'_..' 31%} “Mr—32%;" T', f"; 3:5,“? I. h - i “'4‘; ’V‘“ 73.2. E ‘ ~ ‘3'“ ' -. Pd] ,.. WE‘J' 1:: ‘ ”:21- m. 5.7 r e- 2, (.79; _J:i._‘:' 4., -~« . 93-7 ..r W 7‘3? aft-1k? :::..,_-.g 2: If... ; “I“, I?“ L Inn‘- "I~'—¥ ‘2- TM: 573. “cit" 2. 3333 2 2223‘ .‘z‘ 23223 333.3% 2' E33233,“ \-l ‘c «I .. m. 2 2 6"" £95“ 1:53;; 1x3; “fir: 'cr ‘3 _ ‘1'"I Iug‘I 9‘: 5.,“ . “I ‘ :12 '“ fig' 22%;: :‘VfigkéEWEt 2m} ‘3‘“ '2‘“ " 532222“ 3%32 “:32 2&3 2132 {2% I3 ? P: A VIII, 2, m .22} $311 335293333133 $525932: 3,33% . . “2""? I 22‘“ ‘Hgv ‘8‘ 2232352 3 . .2 II ‘2?' £33: ‘1‘"32‘131‘3 522%.,52 2‘ 'u‘Q; 13333332 . ‘ MM". . I III“ ‘ ‘ 3:. .1 r31 I I)" 5 a . "“fiintx gig "29235525! 5*" 2.3 ..3 _ ,, :7 .., .4. ..L; «”7 J :g‘sifiqr '51 . ‘ «9 47:53:? .335 3.. 32F .75 -:"?§" ~: ., $50“ 5:" 554.? . 53“ "' nah-r 2 fivaz—g; .=—' .- ufgw‘ far-"'- 4*: ' V's”; 1;: gr: —,7' .153. up) -5715} I 1“ 5‘ r- If“ 1:41 1;” ‘ f , .‘ ,_ a: fix ' ' ‘ dis: 4. "S . ~; 1. .3”? ."§ '3 min-{f M5?“ '1'? :0“ :x’ ' ,.:-~ 9.; ., 55¢; ”45c% J’vfié—J 4:»- _ v’ 3-6-3?! ‘ 9’. -7'-:-I: 1:? $ 3. 55% if... 5:: - €32}? 'EE . 1-? fit; “If 1" . .91” ' ' ‘miqi "J." 7.1"." 7,523? *7". - i”: ”f.- ~14.-. . ‘v‘jsv-‘flwg .Ir “345' 9-23. ‘1]. _ ; J“ '2 51712.: $35.23* ' fiwr g“- : 33.1?" J‘ ‘I' as: '1.4 . yr :23? .5 ":2“: 5”!" ..A .._., fig” 51.- ” 4:5; J'.. .9 -- ‘3 ‘55:}! ”1:77 .1? .1 £5717; 2;.” 1.2% 3—, 47..-, - #fi‘r“? f—é—‘J - :5.- ..” _ 5-”. I 1‘14 2:13;! ‘15:?) ‘5‘; @323, 333.51 2‘" 5 fl" ”‘5 M its?" gfi‘ffié’ 6-1.”; V; .14 €gs;=—’E ; m2 I22 I; q‘sfi‘l ' 1m” WI‘I‘I‘ 3‘33: ~ 31,533 2. a“ 2231' $213 ”Ems 2‘s 2“" ' . 3242» I 3c ‘1 55%“ THESIS _ LIBRARY nichigan State University ' This is to certify that the dissertation entitled The Effect of Quinidine on the Positive Inotropic Actions of Digoxin in Isolated Cardiac Muscie presented by Joshua R. Beriin has been accepted towards fulfillment of the requirements for Ph.D. degfiehl Pharmacoiogy/ ox1co ogy WW fl/Qfl] Major professor T ' Ak February 21, 1985 al era Date_.._..____ MSU [J an Affirmative Anion/Equal Opportunity Instilulion 0-12771 fl—ii MSU RETURNING MATERIALS: Piace in book drop to LjBRARJES remove this checkout from w your record. \FINES W‘iii be charged if book is returned after the date stamped beiow. emu.“ r ‘ 1 3 «C. \‘ THE EFFECT OF QUINIDINE ON THE POSITIVE INOTROPIC ACTIONS OF DIGOXIN IN ISOLATED CARDIAC MUSCLE By Joshua R. Berlin A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology 1985 33a-37?/ ABSTRACT The Effect of Quinidine on the Positive Inotropic Actions of Digoxin in Isolated Cardiac Muscle by Joshua R. Berlin The effect of quinidine-induced elevations in plasma digoxin con- centration on the pharmacologic effects of the glycoside are unclear. One possibility was that quinidine could decrease the specific binding of the glycoside to the putative receptor for digoxin (specific bind— ing), the sarcolemmal sodium pump, because quinidine reduces sodium influx into the cardiac muscle fibers. The present study was under- taken to determine whether quinidine does decrease the amount of sodium available to the sodium pump, thereby reducing the specific binding of digoxin and decreasing the positive inotropic effect of the glycoside in left atrial muscle preparations of guinea—pig heart. Quinidine decreased the rate of sodium influx in atrial muscle as indicated by a frequency-dependent reduction in maximal upstroke velo— city (dV/dtmax 86Rb+ uptake in beating heart muscle (86Rb+ uptake). In ) of the action potential and in steady-state ouabain- sensitive preparations stimulated at 3 Hz, 20 uM quinidine decreased 86Rb+ uptake by 30:3% and dV/dt by 50%. Under these conditions, quini- max dine failed to reduce either the rate of onset or magnitude of the positive inotropic effect of 0.6 uM digoxin or the specific binding of Joshua R. Berlin the glycoside. Effects of quinidine were compared with those of benzocaine which decreased dV/dtmax at concentrations 3300 uM. Benzo— 86Rb+ uptake by 32:ll% and delayed the onset caine (300 uM) decreased of the positive inotropic effect of digoxin; however, specific binding of digoxin was not significantly reduced by benzocaine. These results indicate that decreases in sodium influx did not reduce glycoside binding to the sodium pump. The binding of [3H]ouabain to Na+,K+-ATPase, the biochemical correlate of the sodium pump, was inhibited by quinidine only in 2+ and ATP with or concentrations 3900 uM (in the presence of Na+, Mg without K+). Ouabain binding was stimulated by Na+ in concentrations of 35 mM, and quinidine failed to influence Na+-induced stimulation of ouabain binding. The curve representing Na+-induced stimulation of glycoside binding indicates that decreases in sodium influx caused by pharmacologic concentrations of quinidine are likely to reduce glyco- side binding by approximately 5-l0%. This small decrease in digoxin binding is not likely to be pharmacologically significant. Thus, quinidine does not have a significant pharmacodynamic interaction with digoxin. ACKNOWLEDGEMENTS I would like to thank Drs. Tai Akera, Theodore M. Brody, Gregory D. Fink, Joseph R. Hume and Ralph A. Pax for their guidance, friend- ship, and patience. I would also like to thank Eileen Allison, Kevin Stuart, and Todd Gauger for their technical assistance, and Diane Hummel for manuscript preparation. In addition, I have also appre- ciated the friendship and helpful discussions of Yuk—Chow Ng, Paul Stemmer and Richard Kennedy. I especially appreciate the support and love given to me through- out my entire graduate studies by my wife, Julie. 11' TABLE OF CONTENTS Page LIST OF TABLES --------------------------------------------------- vi LIST OF FIGURES -------------------------------------------------- vii INTRODUCTION ----------------------------------------------------- l A. General background ------------------------------------- l B. Effects of quinidine on digoxin pharmacokineticse ------ 2 C. Clinical implications of the quinidine/digoxin inter- action ------------------------------------------------- 8 D. Mechanism of the positive inotropic effect of cardiac glycoside ---------------------------------------------- l2 l. Determinants of glycoside inhibition of the Na+ ,K+ —ATPase enzyme ------------------------------ l2 2. Positive inotropic effect of cardiac glycosides and binding to the Na+,K+-ATPase ------------------ l3 3. Sodium pump inhibition and the positive inotropic effect of cardiac glycosides ---------------------- l6 4. Determinants of glycoside binding to the sodium pump ---------------------------------------------- l7 5. Mechanism of the positive inotropic effect of cardiac glycosides -------------------------------- Zl E. Possible mechanisms of quinidine interaction with digoxin ------------------------------------------------ 22 MATERIALS AND METHODS -------------------------------------------- 28 A. Inotropic studies -------------------------------------- 28 8. Action potential studies ------------------------------- 29 C 86Rb+ uptake studies ----------------------------------- 30 D [3 H]0uabain binding studies ---------------------------- 33 l. Estimation of digoxin binding to Na+ ,K+ -ATPase in heart muscle preparations: Fractional occupancy-- 33 2. [3H]Ouabain binding to guinea pig heart microsomes 34 3. [ 3H]0uabain binding to guinea pig heart Na+ ,K+ - ATPase -------------------------------------------- 36 E. Miscellaneous ------------------------------------------ 36 TABLE OF CONTENTS (continued) RESULTS ---------------------------------------------------------- A. B. Effect of quinidine on the inotropic action of digoxin in atrial muscle preparations of the guinea pig -------- Effect of quinidine and benzocaine on the transmembrane action potential --------------------------------------- l. Quinidine _________________________________________ 2. Benzocaine ________________________________________ Effecgg of quinidine and benzocaine on ouabain-sensi- tive Rb uptake in beating atrial muscle preparations l. Quinidine ......................................... 2. Benzocaine ________________________________________ Inotropic actions of digoxin in the presence of quini- dine or benzocaine ..................................... T. Quinidine ----------------------------------------- 2. Benzocaine ........................................ Digoxin binding in beating heart muscle preparations—-- l. Quinidine ----------------------------------------- 2. Benzocaine ---------------------------------------- 3. Effect of quinidine on the digoxin binding to beating heart muscle incubated in rubidium-con- taining buffer solutions -------------------------- Effect of quinidine on [3H]ouabain binding to Na+,K+— ATPase ------------------------------------------------- l. [3H]Ouabain binding to guinea pig heart microsomes 2. [ H]0uabain binding to partially purified guinea pig heart Na+,K+-ATPase in the presence of low sodium concentrations ----------------------------- DISCUSSION _______________________________________________________ A. Influence of quinidine and benzocaine on the positive inotropic action of digoxin ............................ l. The effect of quinidine on the inotropic action of digoxin ig_vivo ----------------------------------- 2. Effects of quinidine on the positive inotropic actions of digoxin in_vitro ----------------------- 3. Effect of benzocaine on the positive inotropic action of digoxin ifl_vitro ------------------------ iv Page 37 37 42 42 48 51 52 57 6O 60 64 65 67 7O 71 74 74 80 84 84 84 86 92 TABLE OF CONTENTS (continued) Page B. Possible eXplanations for the lack of effect of quini- dine and benzocaine on digoxin binding ----------------- 94 l. Effect of quinidine and benzocaine on sodium in- flux into atrial muscle preparations -------------- 95 a. Measurement of sodium influx ----------------- 95 b. Effect of quinidine and benzocaine on sodium influx rate ---------------------------------- lOl 2. Effect of the positive inotropic action of quini— dine on digoxin binding --------------------------- l07 3. Effect of quinidine on the fractional occupancy of digoxin in rubidium-containing buffer solutions l09 4. Quinidine-induced stimulation of digoxin binding to Na+,K+-ATPase ---------------------------------- llO 5. Sodium dependence of glycoside binding to the Na+,K+-ATPase ------------------------------------- lll BIBLIOGRAPHY——---—4 ---------------------------------------------- ll7 LIST OF TABLES Table Page l Effect of quinidine on fractional occupancy by digoxin in beating left atrial muscle -------------------------- 69 2 Effect of benzocaine on digoxin fractional occupancy in guinea pig left atrial muscle ----------------------- 7l 3 Effect of quinidine on digoxin fractional occupancy in atrial muscle incubated in Rb+-containing buffer solu— tion --------------------------------------------------- 75 vi Figure 10 ll 12 l3 l4 LIST OF FIGURES Page Effect of stimulation frequency on 86Rb+ uptake -------- 32 Effect of quinidine on developed tension of guinea pig left atrial muscle ------------------------------------- 39 Effect of quinidine on the time course of the positive inotropic action of 0.6 uM digoxin --------------------- 4l Effect of quinidine on transmembrane action potential—- 44 Effect of quinidine on atrial muscle action potential at various stimulation frequencies --------------------- 46 Effect of benzocaine on the transmembrane action poten- tial --------------------------------------------------- 49 Effect of benzocaine on transmembrane action potential- 50 Effect of quinidine on ouabain-sensitive 86Rb+ uptake-— 53 Effect of 20 uM quinidine on ouabain-sensitive 86Rb+ uptake at different stimulation frequencies ------------ 55 86 Effect of benzocaine on ouabain-sensitive Rb+ uptake- 58 Effects of quinidine on the time course of the inotro- pic actions of digoxin in atrial muscle preparations stimulated at 0.5 and 3.0 Hz --------------------------- 62 Effect of 20 uM quinidine on the positive inotropic effect of digoxin at different stimulation frequencies- 63 Effect of benzocaine on the positive inotropic action of digoxin --------------------------------------------- 66 Effect of K+ on ATP—dependent [3H]ouabain binding to cardiac microsomes ------------------------------------- 76 vii LIST OF FIGURES (continued) Figure Page l5 Effect of quinidine on ATP-dependent [3H]ouabain binding to cardiac microsomes -------------------------- 79 l6 Effect of sodium ion on ATP-dependent [3H]ouabain binding ------------------------------------------------ 82 viii INTRODUCTION A. General Background Digitalis glycosides and quinidine have been administered concomi- tantly for the treatment of cardiac rhythm disorders. As early as l932, however, Gold et_al, (T932) cautioned against the hazards of this drug combination. Using the database collected by the Boston Colla- borative Drug Surveillance Program, retrospective studies have shown that the incidence of adverse drug reactions in patients on digoxin maintenance therapy was increased after the inclusion of quinidine in the patient drug regimen (Cody et_al,, 1980) and that the majority of patients with apparent quinidine-induced ventricular dysrhythmias were also receiving digitalis glycosides (Cohen §t_gl,, l977). These find— ings gave suggestive evidence for a drug interaction between quinidine and digoxin. The incidence of untoward effects for quinidine alone was almost 30% (Bigger and Hoffman, l980), so that it was not clear if the increase in the number of adverse drug reactions was simply a result of an additive drug effect. In 1978, three laboratories (Doering and Konig, l978; Ejvinsson, l978; Leahey gt 31., l978) independently re- ported that quinidine increased serum digoxin concentration when the antiarrhythmic agent was administered to patients on digoxin mainte- nance therapy. These reports have stimulated a great deal of research designed to investigate the mechanism by which quinidine increases 2 serum digoxin levels and the clinical implications of the increased digitalis levels on cardiac function. B. Effects of Quinidine on Digoxin Pharmacokinetics Reports subsequent to those in l978 have shown that quinidine, in therapeutic doses, produces an approximately l00% increase in digoxin levels in the serum of patients and healthy volunteers receiving digoxin (Bigger, l979; Fichtl and Doering, l983). The increase in serum digoxin levels, though quite variable between individuals, appears to be proportional to the dose of quinidine administered (Doering and Konig, 1978; Powell et_al,, l980). Serum digoxin concentration begins to increase within 24 hr of initiation of quinidine, reaches a plateau in 4 to 5 days and remains elevated for the duration of quini- dine administration (Leahey et_al,, l978; Doering, l979; Doering and Konig, l98l; Leahey gt_al,, l98l). Quinidine could increase serum digoxin concentration by several mechanisms. Alterations in digoxin absorption, distribution and/or elimination might be responsible for the elevation in digitalis levels. As will become apparent, the actions of quinidine on digoxin pharma- cokinetics are quite complex and are, as yet, not completely understood. Chen and Friedman (1980) have shown that a single dose of quini- dine enhanced absorption of digoxin from the gastrointestinal tract. Digoxin, however, is well absorbed after oral administration (>70%; Fichtl and Doering, l983) so that even a marked increase in the adsorp- tion of the administered dose would not account for the magnitude of the change in serum digoxin concentration seen in the presence of quinidine. Furthermore, B-methyldigoxin, a lipophilic analogue of 3 digoxin which is absorbed with almost 100% efficiency from the gut, exhibits similar increases in serum concentration as digoxin with concomitant quinidine administration (Doering, 1979). Therefore, changes in absorption of digoxin do not appear to explain the marked increase in serum digoxin observed with combined drug therapy. A decrease in the elimination rate of digoxin from the body could account for the sustained rise in serum glycoside concentration. The rate of disappearance of digoxin from plasma was slowed by quinidine in volunteers administered an i.v. bolus dose of digoxin (Steiness et_al,, 1980). Similar results were reported in patients on combined digoxin/ quinidine therapy (Doering et_al,, 1981; Pedersen §t_al,, l983). Conflicting data, though, has been reported by Hager et_al, (1979) who found no consistent effect of quinidine on the half-time of the dis- appearance of digoxin from plasma. Even so, serum elimination half— times increased in 3 of 6 individuals. Quinidine appeared to decrease the rate of elimination of digoxin from the body. Digoxin is eliminated from the body primarily by the kidney (Hoff- man and Bigger, 1980), and several studies have shown that renal clear- ance of digoxin is decreased by quinidine (Doering, 1979; Hager et_al,, 1979; Steiness et_al:, 1980; Leahey et 31,, 1981). Creatinine clear- ance, an index of glomerular filtration rate (Doering, 1979; Hager gt 91:, 1979; Steiness et_al,, 1980), and plasma protein binding (Fichtl and Doering, 1983) of digoxin are unchanged by quinidine. Taken to- gether these data show decreased renal clearance of digoxin observed during concomitant quinidine administration is due to a direct effect of quinidine on the renal handling of digoxin. Leahey 23.91: (1981) found that the magnitude of the decrease of renal digoxin clearance 4 correlated well with the plasma quinidine concentration, even at levels below therapeutic quinidine concentrations. These results suggest that altered renal elimination of digoxin may, in part, be responsible for elevated serum digoxin levels seen in patients on combined quinidine/ digoxin therapy. Extrarenal clearance which accounts for approximately one-third of total elimination of digoxin (Hoffman and Bigger, 1980), is decreased by quinidine to a similar extent as renal clearance (Steiness et_al:, 1980; Pedersen §t_al,, 1983). Quinidine also elevates plasma digoxin concentration in anuric patients on chronic digoxin maintenance therapy to a similar extent as seen in patients with normal kidney function. Serum digoxin in these patients reached a new steady-state concentra- tion 8 days after initiating quinidine administration, whereas serum quinidine concentration remained constant after 3 days (Fichtl gt al,, 1983). That serum digoxin continued to rise after serum quinidine con— centration had plateaued indicates that the increase in the concentra- tion of digoxin was not simply due to a competitive displacement of digoxin from tissue binding sites by quinidine. Fichtl et al, (1983) concluded from these findings that extrarenal clearance of digoxin was slowed although values for digoxin clearance were not reported. Quinidine decreases the rate of digoxin elimination from the body both by renal and extrarenal pathways, and a recent overview of published reports (Fichtl and Doering, 1983) points out that the magnitude of the observed decrease in digoxin clearance (>40% reduction) produced by quinidine may be sufficient to account for the doubling of serum di- goxin concentration reported in patients simultaneously administered quinidine and digoxin. 5 Alterations in digoxin clearance from the body may not be the sole mechanism by which quinidine increases serum glycoside concentration. Leahey et_al, (1978) noted that serum digoxin concentration was in- creased by quinidine even when digoxin administration was discontinued prior to the initiation of quinidine therapy. Risler et_al, (1980) reported, in contrast to Leahey et_al, (1981), that although the in- crease in serum digoxin concentration depended on the administered dose of quinidine, renal clearance of digoxin was decreased to a similar extent whether quinidine was given in a low (500 mg/day) or high (1000 mg/day) dose.i These results suggested that decreased renal clearance was not the major cause for elevated serum digoxin concentration. Displacement of tissue stores of digoxin was suggested by the authors as the underlying cause for these effects. Numerous studies, have since been undertaken to examine the influence of quinidine on digoxin distribution within the body. An increase in plasma digoxin concentration may result from a decline in the apparent volume of distribution of the drug. Published reports dealing with the influence of quinidine on the distribution volume of digoxin after intravenous administration of the glycoside have yielded conflicting results. No change (Steiness et_al,, 1980; Leahey gt al,, 1981) or a decrease (Hager et_al,, 1979) in the volume of digoxin distribution has been noted. Aside from differences in methodology, the conflicting result may be due to differences in the administered dose of quinidine. Both Steiness et_alf (1980) and Leahey et_al, (1981) administered quinidine which resulted in mean serum quinidine concentrations of less than 2 ug/ml while the mean serum concentration achieved in the study of Hager et_al, (1979) was 2.5 6 ug/ml. Indeed, when Leahey et_al, (1981) segregated the subjects into low (less than 1.9 ug/ml) and high (greater than 1.9 ug/ml) serum quinidine concentration,.a significant decrease in apparent volume of distribution was observed in those subjects with "high" serum quinidine concentrations. A recent study by Pedersen gt_al, (1983) in patients on chronic quinidine/digoxin therapy verifies the results of the earlier single dose studies. The volume of distribution of digoxin was reduced in patients on quinidine as compared to when quinidine therapy was withdrawn. The magnitude of the decrease was correlated with the plasma quinidine concentration. These data suggest that quinidine decreases the volume of digoxin distribution in patients on combined quinidine/digoxin therapy. A quinidine-induced mobilization of tissue stores of digoxin may explain the decreased volume of digoxin distribution. As already noted, the presence of quinidine does not influence the binding of digoxin to plasma proteins. Schenck-Gustafsson gt_al, (1981) reported that, although skeletal muscle digoxin content increased 50% as com— pared to before quinidine administration, the ratio of skeletal muscle to serum digoxin concentrations decreased by 23% in patients treated for atrial fibrillation after 4 days of concomitant quinidine admini- stration. This was taken as evidence for reduced binding of digoxin to skeletal muscle. Similar evidence, in dogs, showed that tissue/serum ratios of digoxin were reduced during quinidine administration in many organs including skeletal and heart muscle, kidney and liver (Doherty et_al,, 1980). The reported decrease in the apparent volume of digoxin distribution may be the result of decreased tissUe binding by digoxin in the presence of quinidine. 7 Changes in digoxin volume of distribution may account for a tran- sient increase in serum digoxin concentration caused by quinidine but cannot explain a sustained increase in serum digoxin concentration. A sustained increase in serum digoxin is possible only if the rate of digoxin elimination is slowed. Changes in volume of distribution will only lead to transient changes in plasma drug concentration (Lee, 1980; Fichtl and Doering, l983). Quinidine produces a sustained increase in serum digoxin concentration (Leahey gt_al,, l978; Doering, 1979) so that a slowing of digoxin elimination itself would be sufficient to produce an increase in serum digoxin. The effect of changes in digoxin distribution, however, should not be discounted as a factor which might contribute to the rise in digoxin concentration immediately after quinidine administration. This notion is supported by studies in animal models which demonstrate that differences in serum digoxin concentration in the presence of quinidine become apparent within 2 hours (Gibson and Nelson, 1979; Kim et_al,, 1981a) when changes in elimination would have only a small impact on serum digoxin levels. In summary, quinidine has complex effects on digoxin pharmaco- kinetics. In the presence of quinidine, serum digoxin concentration is elevated in patients receiving digoxin maintenance therapy. Quinidine appears to decrease the volume of distribution of digoxin within the body which may result in a transient increase in serum digoxin concen- tration. The sustained rise in digoxin levels must be due to a de- crease in the rate of elimination of digoxin, and pharmacokinetic studies suggest that both renal and extrarenal clearance of digoxin is decreased. 8 C. Clinical Implications of the Quinidine/Digoxin Interaction From a clinical perspective, the most important point to be re— solved is whether the quinidine-induced elevation in serum digoxin concentration results in a comparable increase in the effect of digoxin on heart function. Studies of the quinidine/digoxin interaction in man and in experimental models have not led to concensus on this point. In a series of restrospective and prospective case studies, Leahey and coworkers looked for evidence of increased glycoside effects in digitalized patients receiving quinidine. In their original paper (Leahey et_al,, 1978), 7 of 27 patients had ventricular arrhythmias develop or worsen after beginning quinidine administration. These adverse reactions and the incidence of nausea were lessened by de- creasing the digoxin dose. Their subsequent studies confirmed the findings in the first paper. Patients on digoxin therapy, when admini- stered quinidine simultaneously, had a lengthening of the PR interval of the electrocardiogram sometimes leading to A—V dissociation, a sign of increased glycoside action, as well as a high incidence of gastro- intestinal distress. These changes were usually noted within 72 hr of initiating quinidine administration (Leahey et_al,, 1979; l980b). Thus, increased serum digoxin concentration was associated with a greater incidence of untoward reactions and ECG evidence of a greater digoxin effect. These studies, however, did not directly address the question of whether the increased serum digoxin concentration was having a greater influence on ventricular performance. Although a greater incidence of ventricular arrhythmias and nausea in patients on quinidine/digoxin therapy might indicate a more pronounced effect of increased serum digoxin, quinidine itself has a high incidence of side 9 effects including arrhythmias and gastrointestinal disturbances (Cohen et 21,, 1977) so that the increased incidence of these symptoms might only reflect the presence of quindine. The increased prolongation of the P—R interval of the ECG, while indicative of a greater glycoside effect, is a measure of digoxin's anticholinergic action, not the drug's effect on ventricular contractility. Thus, these studies, while suggestive of an increased effect of digoxin in patients receiving both quinidine and digoxin, did not demonstrate an increased effect on cardiac performance. Subsequent studies in man, examining the effects of digoxin on cardiac performance, are divided on whether increased serum glycoside concentration correlates with an increased effect on the heart. Some studies have demonstrated that the effect of digoxin on cardiac con- tractility was reduced after concomitant quinidine administration was begun. Cardiac contractility was measured indirectly by determining the effect of various drug regimens on the electromechanical systole of the heart. Electromechanical systole or the systolic time interval (511) is shortened by inotropic concentrations of digoxin. Two studies demonstrated that the shortening of STI by digoxin was reduced after beginning quinidine administration (Hirsch et al,, 1980; Steiness at al,, 1980), which indicated that the inotropic effect of digoxin was decreased. These studies have been criticized, however, in that Steiness et_al, (1980) did not report the effect of quinidine alone on STI and Hirsh et_§l, (1980) measured the effects of quinidine in less than half of the cases receiving digoxin and quinidine, though in those individuals studied, quinidine did not produce a significant effect on STI. Belz et_al, (1982), conducting a double-blind, crossover study, 10 found that the effect of digoxin on STI shortening in volunteers re- ceiving concomitant quinidine and digoxin was not greater than when they received digoxin alone; however, quinidine itself produced a significant lengthening of STI. Thus, the effect of digoxin on STI was greater with simultaneous quinidine administration when the effect of quinidine was taken into account. This finding was confirmed by ele- vating the serum digoxin concentration in the absence of quinidine to the same level as that seen with quinidine/digoxin administration. In both cases, STI were shortened to a similar extent by digoxin. Zaman et a1, (1981), using echocardiographic techniques, also found evidence for an increased effect of digoxin on the rate of ventricular shorten- ing when quinidine was administered simultaneously. These studies demonstrated that the increase in serum digoxin caused by quinidine was correlated with an enhanced glycoside effect on cardiac function, in opposition to the earlier studies (Hirsh gt alt, 1980; Steiness et al:, 1980) which found no such correlation. The differences in these in— vestigations may have resulted from the different protocols used. Investigations using animal models of the quinidine/digoxin inter- action have also given conflicting results concerning the effects of elevated serum digoxin levels. Kim et_al,‘(l981b) found that occupancy of cardiac Na+,K+-ATPase, the putative pharmacologic receptor for cardiac glycosides (Akera, 1981), was increased in animals co-admini- stered quinidine and digoxin at a time when serum digoxin concentration was significantly greater than in animals administered digoxin only. In dogs, elevated serum digoxin produced by concomitant quinidine administration was shown to have a greater pharmacologic effect as measured by inhibition of active monovalent cation transport (Leahey et_ 11 21:, l980a), which is proposed as the mechanism of glycoside action in the heart (Akera, 1981). In agreement with the above studies, acetyl— strophanthidin tolerance (AcST) tests in dogs showed that digoxin had a greater pharmacologic effect when its serum concentration was elevated by quinidine (Wilkerson 22.21:, 1984). The AcST test uses the inverse relationship between the pharmacologic effect of digoxin and the dose of acetylstrophanthin to produce premature ventricular beats to deter- mine the degree of digoxin's pharmacologic effect on the heart. The test is complicated because quinidine alone increases AcST; however, when the effect of quinidine is considered, the reduction in AcST produced by digoxin in animals receiving both quinidine and digoxin is greater than the reduction in AcST produced by digoxin in dogs admini- stered the same dose of digoxin only (Wilkerson et_al,, 1984). One study (Warner et_al,, 1984), though, shows a dissociation between the serum concentration of digoxin and its pharmacologic actions in dogs co-administered quinidine. In this study, quinidine increased serum digoxin levels 100 percent, but the inhibition of active monovalent cation transport in cardiac ventricular muscle from these dogs was the same as the inhibition of cation transport seen in cardiac muscle taken from dogs receiving digoxin only. Furthermore, when dogs on digoxin 9nly_were titrated to a serum glycoside concentration the same as that seen in animals receiving both quinidine and digoxin, the expected increase in inhibition of active cation transport was observed. Al— though the serum digoxin concentration was the same in both groups, the effect of digoxin on monovalent ion transport was significantly less in dogs administered quinidine. The reasons for the conflicting results in these studies is not clear. Thus, the possibility that quinidine interferes with the pharmacologic actions of digoxin remains unsettled. 12 D. Mechanism of the Positive Inotrgpjc Effect of Cardiac Glycosides Cardiac glycosides have been known to inhibit monovalent cation transport across cellular membranes for several decades (Schatzmann, 1957; Post et_al,, 1960). Thus, when a sodium, potassium-activated adenosine triphosphatase activity was identified in the microsomal fraction of crab neurons (Skou, 1957) and in other tissues (Hess and Pope, 1957; Post et_al,, 1960), the activity of which was blocked by cardiac glycosides, the hypothesis was soon advanced that the positive inotropic effect of the glycosides was due to inhibition of the cardiac sarcolemmal Na+,K+-ATPase (Repke, 1963; Glynn, 1964). 1. Determinants of glycoside inhibition of the Na+,K+-ATPase enzyme Binding of cardiac glycosides to isolated Na+,K+-ATPase enzyme requires the presence of Mgz+ and is stimulated by ATP or inor- ganic phosphate (Matsui and Schwartz, 1968; Schwartz gt_al,, 1968). The binding of digitalis is further enhanced by Na+ but is antagonized by K+ in the presence of Mg2+ 2+ and ATP, whereas, binding in the presence of Mg and Pi is inhibited by both Na+ and K+ (Matsui and Schwartz, 1968; Schwartz et_al,, 1968; Akera and Brody, 1971). The ligand requirements for glycoside binding appear to be the same as those required for formation of the phosphorylated intermediate of the native Na+,K+-ATPase. Sodium, in combination with Mg2+ and ATP, promotes the formation of the phosphoenzyme and potassium enhances the rate of dephosphorylation of the enzyme (Sen gt_a1,, 1969; Skou and Hilberg, 1969). Inorganic phosphate also phosphorylates the enzyme in the presence of Mg2+ (Lindenmayer gt 31,, 1968; Sen §t_al,, 1969). Prefer— ential binding of the glycoside, thus, appears to occur to a 13 phosphorylated conformation of the enzyme, EZ-P (Matsui and Schwartz, 1968; Sen 3331., 1969; Tobin gig-$1., 1972). Although ouabain and 1d compete for binding to the same enzyme conformation, the glycoside binding site is distinct from that for K+ (Matsui and Schwartz, 1966) and is located on the outer surface of the intact membrane (Perrone and Blostein, 1973). Cardiac glycoside binding to the Na+,K+-ATPase leads to a stabilization of the phosphoenzyme (Lindenmayer et_al,, 1968; Sen et al,, 1969) and further binding of ATP is blocked (Sen et_al,, 1969; Skou and Hilberg, 1969). Consequently, binding of cardiac glycosides correlates closely with inhibition of Na+,KI—ATPase activity, and, in intact cellular preparations, binding parallels drug-induced inhibition of active Na+ and K+ fluxes across the cell membrane of erythrocytes (Hoffman, 1969; Gardner and Kiino, 1973) and squid axon membranes (Baker and Willis, 1972). Recently, some cardiac muscle membranes have been shown to have two classes of glycoside binding sites, but only one class of sites, the low affinity binding sites, have been associated with inhibition of Na+,K+-ATPase activity and inhibition of active cation flux (Kazazoglou §t_al,, 1983). Thus, cardiac glycosides speci- fically bind to the Na+,K+-ATPase and binding is modulated by the presence of Na+ and K+. The consequences of binding are inhibition of Na+,K+-ATPase activity and active monovalent cation transport in intact cells. 2. Positive inotropic effect of cardiac glycosides and binding to the Na+,K+-ATPase The positive inotropic action of cardiac glycosides in the heart is believed to result from partial inhibition of the cardiac 14 sarcolemmal Na+,K+-ATPase. This hypothesis has been supported by several lines of evidence which show that cardiac glycosides bind to the Na+,K+—ATPase and inhibit monovalent cation transport at the time of their positive inotropic action in the heart. The Na+,K+-ATPase isolated from cardiac ventricular muscle of anesthetized dog admini- stered ouabain has been shown to be inhibited to a degree related to the positive inotropic effect of the drug (Akera §t_al,, 1970; Allen at al,, 1975). Similar results were obtained in isolated in situ dog heart preparations (Besch et_a1,, 1970). These studies have been supported by the finding that the glycoside is bound to the Na+,K+- ATPase at the time of the positive inotropic effect in isolated heart muscle preparations (Ku et_al,, 1974; Schwartz et_al,, 1974). Binding of the glycoside is also concentration-dependent (Allen et 31,, 1975) and the degree of binding correlates well with the onset and washout of drug effects (Ku et 31,, 1974; Schwartz et a1,, 1974). Furthermore, the function of other cellular components, the mitochondria and sarco— plasmic reticulum, which might alter cardiac muscle contractility, were not changed during cardiac glycoside inotropic action (Besch §t_a1:, 1970; Allen gt_al,, 1975). Thus, digitalis binds to and is capable of inhibiting the Na+,K+-ATPase at the time of its positive inotropic effect. The characteristics of the glycoside/enzyme complex formed_in ijg_and in heart muscle preparations show similar characteristics as the complex formed with isolated Na+,K+-ATPase. The glycoside/Na+,K+- 2+ and ATP is ATPase complex formed ifl_vitro in the presence of Na+, Mg stabilized in the presence of KCl (Akera and Brody, 1971). Similarly, the dissociation of bound glycoside from tissue homogenates of canine 15 ventricular muscle exposed to ouabain for a period of time before homogenization is slower in the presence of KCl (Akera §£.§l:’ 1976b). Potassium also antagonizes glycoside binding to the isolated Na+,K+- ATPase (Matsui and Schwartz, 1968; Akera and Brody, 1971). Likewise, when extracellular KCl is increased the rate of development of the positive inotropic effect and the increase in tissue content of digoxin in cat ventricular papillary muscles is delayed (Prindle et a1,, 1971). In anesthetized dog, hyperkalemia reduced the positive inotropic effect of digoxin and decreased the drug-induced inhibition of the Na+,K+- ATPase (Goldman et_al,, 1973). i The potency of cardiac glycosides to inhibit the Na+,K+- ATPase varies greatly depending on the source of the enzyme. The concentration of ouabain to produce a positive inotropic effect in isolated guinea pig and rat cardiac muscle correlates well with the concentration of drug required to inhibit the Na+,K+-ATPase isolated from guinea pig and rat heart. In both cases, the rat required an approximately 100 times greater drug concentration (Ku et_al,, 1976). Similar species-dependent effects of cardiac glycosides have been obserVed in dog, cat, and rabbit (Akera §t_al,, 1973). In all cases, the concentration of digitalis which inhibits the activity of the cardiac Na+,K+-ATPase correlated well with the concentration to produce a positive inotropic effect. These studies, taken together, demon- strate that cardiac glycosides bound to the Na+,K+-ATPase during their positive inotropic action display characteristics similar to the drug bound to the Na+,K+-ATPase jn_vitrg, which suggests that glycoside binding is closely related to the positive inotropic effect of these drugs. 16 3. Sodium pump inhibition and the positive inotropic effect of cardiac glycosides The sodium pump, the functional correlate of the Na+,K+- ATPase, maintains the ionic gradients across the cellular membrane against passive leak of ions down their electrochemical gradients. Inhibition of the Na+,K+-ATPase by positive inotropic concentrations of digitalis should inhibit a fraction of the functional sodium units which would lead to an increase in intracellular sodium and a loss of intracellular potassium. Some studies, however, have shown that myo- cardial Na+ and K+ content does not change with inotropic concentra- tions of cardiac glycosides (Lee and Klaus, 1971; Bentfeld et al,, 1977). Akera et_al,(l976a) has postulated that monovalent cation content need not change markedly in the face of moderate sodium pump inhibition, because the “reserve capacity" of the pump, the difference between sodium influx rate and the capacity of the sodium pump to extrude sodium, will still allow the sodium pump to maintain the Na+ and K+ gradients across the membrane. Nevertheless, the magnitude of the transient increase in intracellular sodium which might be expected to occur following membrane excitation would be enhanced by moderate inhibition of the pump. More recent studies demonstrated that sodium content does increase in cultures of chick embryonic heart cells (Biedert et_al,, 1979; Kazazoglou gt_al,, 1983) in a dose-dependent manner for positive inotropic concentrations of digitalis. Further- more, diastolic free intracellular sodiUm concentration, measured with sodium-sensitive microelectrodes, has also been shown to increase after exposure of isolated heart muscle preparations to nontoxic concentra- tions of digitalis (Lee at al,, 1980; Lee and Dagostino, 1982; Wasserstrom 17 et_al,, l983). Alterations in intracellular sodium concentration do occur in the presence of positive inotropic concentrations of cardiac glycosides, a result consistent with sodium pump inhibition. Direct confirmation that the sodium pump is inhibited by digitalis comes from measuring the rate of monovalent cation transport across the sarcolemmal membrane. The maximal rate of monovalent cation transport, the pump capacity, is decreased in a dose-dependent manner at subtoxic glycoside concentrations (Kazazoglou et al,, 1983). The decrease in pump capacity also correlates well with the onset and washout of ouabain's positive inotropic effects in isolated heart muscle (Ku et al,, 1974), and pump activity has also been shown to be decreased during the onset of drug effects (Hougen and Smith, 1978; Biedert gt_al,, 1979). These studies demonstrate that pump inhibition is closely related to the positive inotropic actions of cardiac glyco- sides in the heart. 4. Determinants of glycoside binding to the sodiumgpump_ As already discussed, digitalis binding to the Na+,K+-ATPase is modulated by the action of Na+ and Ki ions to induce the enzyme to take the K+-sensitive phosphoenzyme conformation and to reduce this form of the enzyme, respectively. Similarly, in the intact cell, the binding of the glycosides to the sodium pump is regulated by the con- centration of intracellular sodium and extracellular potassium, pre- sumably reflecting the availability of the glycoside-sensitive confor— mation of the Na+,K+-ATPase. Raising extracellular potassium decreases the rate at which ouabain inhibits active cation transport in squid axon (Baker and Willis, 1972), decreases glycoside-induced inhibition + + . . . of the Na ,K -ATPase measured 1n can1ne cardiac muscle exposed to 18 ouabain (Goldman et_al,, 1973) and lessens the development of the positive inotropic actions of ouabain (Goldman gt_al,, 1973; Prindle at 31,, 1973)° Conversely, decreasing extracellular potassium enhances the rate of glycoside inhibition of active cation transport in squid axon (Baker and Willis, 1972). These studies show that binding of ouabain to the sodium pump is influenced by extracellular potassium which is likely to result from the K+-induced reduction in the avail— ability of glycoside-sensitive phosphoenzyme. Maneuvers which increase intracellular sodium, on the other hand, promote glycoside binding to the sodium pump. Several studies (Moran, 1967; Park and Vincenzi, 1975; Bentfeld et_al,, 1977) have shown that the development of glycoside actions at different rates of electrical stimulation is dependent on the nUmber of contractions not on the duration of glycoside exposure. This "beat dependency" of glycoside action was further shown to depend on the number of membrane depolarizations rather than the number of contractions (Akera et_al,, 1977). The rate of ouabain binding (Yamamoto §t_al,, 1979; lemma and Akera, 1982) and inhibition of active monovalent cation transport (Yamamoto et_al,, 1979) are also increased by higher stimulation fre- quencies. These experiments demonstrate that the onset of cardiac glycoside action is highly dependent on the frequency of electrical stimulation. Increasing stimulation rates have been shown to increase diastolic intracellular sodium concentration (Cohen §t_al,, 1982; January and Fozzard, 1984) and increase sodium pump activity (Yamamoto et_al:, 1979; Akera et 31,, 1981). Compounds which increase sodium influx rate such as monensin, a sodium ionophore (Meier et_a1,, 1976) l9 and the sodium channel toxins, batrachatoxin (Albuquerque §t_al,, 1973) and grayanotoxin (Narahashi and Seyama, 1974) which increase sodium influx via the sodium channel, increase the rate of development of the positive inotropic effect of ouabain (Akera gt_al,, 1977) and enhance glycoside binding (Temma and Akera, 1982) in isolated cardiac muscle preparations. Thus, increased sodium transport by the sodium pump enhances the rate of digitalis action; again, presumedly by in- creasing the availability of the glycoside-sensitive conformation of the Na+,K+-ATPase. The effects of lowering intracellular sodium on cardiac glycoside action are not as easily explained solely by postulating that glycoside binding is dependent on the availability of the K+-sensitive phosphoenzyme conformation of the Na+,K+-ATPase. Isolated cardiac muscle preparations incubated in a low sodium (85 mM) salt solution showed a delayed onset of the positive inotropic action of ouabain when compared to muscles bathed in normal sodium (145 mM) solution (Akera_et al,, 1977). Lowering extracellular sodium concentration to 85 mM has been shown to decrease resting intracellular sodium activity by approxi- mately 30% in sheep heart Purkinje fibers (Ellis, 1977). Incubating cardiac muscle preparations in sodium depleted (20-30 mM) or sodium- free buffers which decreases free intracellular sodium concentration more than 70% (Ellis, 1977), prevents the positive inotropic effect of cardiac glycosides (Linden and Brooker, 1980; Wiggins and Bentolila, 1980; Temma and Akera, 1983). Although Linden and Brooker (1980) believed that ouabain binding still occurred under these conditions, Temma and Akera (1983) demonstrated that ouabain binding to the Na+,K+- ATPase did not occur in guinea pig left atrial muscle exposed to the 20 glycoside in a 27 mM Na+ buffer solution. These data are in agreement with the postulate that intracellular sodium ion concentration modu— lates glycoside binding to the sodium pump; however, other data appear to be at odds with this postulate. Inactivating the sodium channels by raising extracellular potassium concentration from 4 to 25 mM decreases free intracellular sodium concentration approximately 20% (Ellis, 1977), yet in cardiac muscle restored to excitability with isoprotere- nol, cardiac glycosides still have a positive inotropic effect (Besch and Watanabe, 1978). The magnitude and time course of the positive inotropic effect of ouabain observed in cat ventricular papillary muscle incubated in 22 mM K+—containing buffer solutions is not signi- ficantly different than that seen in preparations incubated in 8 mM K+ buffer solutions (Wiggins and Bentolila, 1980). The effect of iso- proterenol on intracellular Na+ concentration in the presence of high extracellular K+ has not been examined. Moreover, epinephrine has been shown to stimulate glycoside binding to the Na+,K+—ATPase in rat soleus muscle immediately after addition of the catecholamine to the incubation medium (Clausen and Hansen, 1977). Therefore, the lack of an effect of high extracellular potassium on the inotropic actions of ouabain may not necessarily indicate that a reduction in intracellular sodium failed to affect the onset of the glycoside action. Isolated cardiac muscle preparations exposed to 3 uM tetrodotoxin, however, still show an increase in contractility in the presence of cardiac glycosides whose rate of development is not significantly different from that seen in the absence of tetrodotoxin (Wassermann and Holland, 1969), even though 3 uM tetrodotoxin is reported to decrease intra— cellular sodium by approximately ten percent (Deitmer and Ellis, 1980a). 21 The moderate decrease in free intracellular sodium ion concentration caused by tetrodotoxin may not be sufficient to produce a significant change in inotropic effect of the glycoside, unlike the more pronounced decrease in intracellular sodium ion concentration (30%) produced by lowering extracellular sodium to 85 mM. Alternatively, changing extra- cellular sodium ion concentration may affect digitalis binding to the Na+,K+-ATPase at an extracellular site in addition to its effect on intracellular sodium. In summary, increased intracellular sodium ion concentration promotes cardiac glycoside binding to the Na+,K+-ATPase; however, the effect of a decrease in intracellular sodium ion concen- tration on the binding and on the positive inotropic actions of digi- talis are not as well established. 5. Mechanism of the positive inotropic effect of cardiac glyco— sides Cardiac glycosides bind to the sarcolemmal Na+,K+-ATPase and, in the intact cell, inhibit the sodium pump producing a transient and/or sustained increase in intracellular sodium during the cardiac contraction cycle. The positive inotropic effect of the glycosides is mediated by an increased intracellular calcium transient during the cardiac action potential (Morgan and Blinks, 1982). The mechanism believed to link increased intracellular sodium to an increased intra- cellular calcium concentration is Na+/Ca2+ exchange across the sarco- lemma (Reiter and Seitz, 1968). Intracellular sodium is coupled to intracellular calcium by the electrochemical gradients for sodium and calcium across the sarcolemma (Mullins, 1979) so that an increase in intracellular sodium is translated into an increase in intracellular calcium by greater calcium influx across the sarcolemma (Glitsch et_ a1,, 1970; Ellis, 1977). Indeed, recent studies using cardiac cell 22 cultures suggest that calcium influx is increased by inotropic concen- trations of cardiac glycosides (Biedert et al,, 1973; Kazazoglou et 31,, 1983) and by veratridine, an agent which increases intracellular sodium independent of sodium pump inhibition (Fosset et_al,, 1977; Pang and Sperelakis, 1982). Thus, the hypothesis has been developed which states that digitalis inhibition of the Na+,K+—ATPase decreases the rate of sodium extrusion by the sarcolemmal sodium pump. The resulting increase in intracellular sodium augments net calcium influx during the action potential via sodium/calcium exchange. The increased calcium influx is manifested as the positive inotropic effect of the cardiac glycosides. E. Possible Mechanisms of Quinidine Interaction with Digoxin From the previous discussion, it is evident that quinidine could interfere with the inotropic actions of digoxin by several mechanisms. Quinidine could inhibit the binding of ouabain to the sodium pump either by competing for the glycoside receptor site of the Na+,K+- ATPase or by decreasing the availability of the glycoside sensitive conformation of the sodium pump. Alternatively, quinidine might inter— fere with the consequences of glycoside inhibition of the sodium pump. Quinidine has been shown to affect many of the subcellular pro— cesses which control calcium metabolism in the heart. Quinidine binds to (Besch and Watanabe, 1977) and inhibits Ca2+ sequestration by iso- lated cardiac sarcoplasmic reticulum vesicles (Fuchs et_al,, 1968; Besch and Watanabe, 1977). Furthermore, quinidine is reported to inhibit membrane currents associated with Na/Ca exchange in frog heart 23 cells (Mentrard et_al,, 1984). Although inhibition of cardiac sarco- plasmic reticulum function or inhibition of sarcolemmal Na/Ca exchange might attenuate the effect of digoxin binding to the Na+,K+-ATPase in the heart, the quinidine concentrations (>10'4M) which were effective at inhibiting these processes were 1-2 orders of magnitude greater than the therapeutic range (3-6 uM) of quinidine. Consequently, quinidine would not appear to interfere with the consequences of sodium pump inhibition by digoxin. Quinidine might antagonize the inotropic actions of digoxin by decreasing binding of digoxin to the Na+,K+—ATPase in the heart. Displacing digoxin bound to its receptor site or preventing glycoside binding to the cardiac muscle Na+,K+-ATPase would antagonize inotropic actions of digoxin. Quinidine has been found to inhibit the activity of the Na+,K+-ATPase isolated from heart (Lowry et_al,, 1973; Besch and Watanabe, 1977) and brain (Lowry et_al,, 1973) as well as inhibit active monovalent cation transport in erythrocytes (Lowry gt al,, 1973; Ball et_al,, 1981). The concentration needed to produce fifty percent inhibition, however, was approximately 1 millimolar. Quinidine has been reported to decrease ouabain binding to the Na+,K+-ATPase. The drug (10"4 - 10'3M) decreased the number of ouabain binding sites in beef heart sarcolemmal preparations without changing receptor site affinity (Straub et_al,, 1978). Another study (Ball et_al,, 1981), however, found that Bmax for glycoside binding remained constant but that the apparent KD increased due to a decrease in the association rate constant. Again, quinidine concentration was close to or equal to one millimolar. Quinidine inhibited Na+,K+-ATPase activity and de- creased glycoside binding to the enzyme but only at very high 24 concentrations. Doering (1979), on the other hand, could not demon- strate any effect of lower concentrations (10-6 — 10-4) of quinidine on ouabain binding to sheep heart Na+,K+-ATPase. It is interesting to note that even in the presence of high concentrations (millimolar) of quinidine, specifically-bound digoxin was not displaced from its recep- tor (Ball et_al., 1981). Thus, quinidine is thought to act at a site distinct from the glycoside receptor -- a proposal supported by the data of Lowry et_a1, (1973) which showed the characteristics of quini- dine inhibition of the Na+,K+-ATPase differed significantly from those of cardiac glycosides. The above reports, therefore, indicate that quinidine is capable of inhibiting digoxin binding to the Na+,K+- ATPase, but the concentrations required were much greater than those used therapeutically. Quinidine could also inhibit digoxin binding by decreasing the availability of the glycoside-sensitive form of the sodium pump indirec- tly rather than competing at the glycoside receptor site. As discussed previously, the availability of the glycoside-sensitive form of the sodium pump is regulated by the levels of extracellular potassium and intracellular sodium. Quinidine probably has little effect on the effective extracellular potassium concentration (Note: Colatsky (1982) has demonstrated that quinidine inhibits the delayed rectifier current in rabbit Purkinje fibers, an effect which might influence extracellu- lar potassium in the beating heart muscle). On the other hand, quini- dine may well have significant effects on intracellular sodium concen- tration. Quinidine has local anesthetic actions and its antiarrhythmic effect has been attributed to its depression of the sodium current in heart muscle. Quinidine produces a significant decrease in the rate of 25 phase 0 depolarization of transmembrane action potentials recorded in cardiac atrial (Vaughan-Williams, 1958) and ventricular muscle fibers (Johnson, 1956; Johnson and McKinnon, 1957). The decrease in upstroke velocity reflects a decline in the rate of sodium entry through the sodium channel into the cardiac cell. The rate of sodium entry measured eitherby isotopic sodium flux (Choi et_al,, 1972) or indirect- ly, by the rate of potassium efflux (van Zwieten, 1969; Klein et_al,, 1960; Choi et_al,, 1972) is decreased by quinidine in heart muscle preparations. A similar decrease in sodium influx rate is believed to be responsible for the decrease in intracellular sodium ion activity produced by the local anesthetic agents, procaine and lidocaine, and by tetrodotoxin in Cardiac Purkinje fibers (Deitmer and Ellis, 1980a; January and Fozzard, 1984). The effect of quinidine on intracellular free sodium ion concentration has not been investigated but it is expected to be similar to the other local anesthetic agents. A quinidine-induced decrease in sodium influx and intracellular sodium ion activity in cardiac tissue might be expected to reduce the binding of cardiac glycosides to the myocardial sodium pump because of a decline in the digitalis-sensitive conformation of the pump. Studies in man have given conflicting results as to whether quinidine actually does decrease the inotropic actions of digoxin. Investigations using animal models of the quinidine/digoxin interaction also yield conflict- ing data on the effects of quinidine on digoxin binding to the cardiac Na+,K+-ATPase and inhibition of active monovalent cation transport. In_ vjvg_studies, however, are complicated by the fact that serum digoxin concentration increases markedly with co—administration of quinidine. 26 It is difficult then to discern whether quinidine does antagonize digoxin effects because the increased serum glycoside concentration may overcome the influence of quinidine. Although the apparent digoxin effect may be increased its effect might be less than in the absence of quinidine. Few studies have thoroughly evaluated this point (but see Belz §t_al,, 1982; Warner gt_al,, 1984). To avoid this complication, the effects of quinidine on the inotropic actions of digoxin have been investigated in isolated heart muscle preparations. In cat ventricular papillary muscle (Williams and Mathew, 1981), prior administration of quinidine decreased the positive inotropic effect of digoxin when compared to the same concentration of digoxin alone. Interestingly, if quinidine was added after digoxin, the net increase in contractility in the presence of both drugs was greater than with the glycoside alone, even though quinidine itself had no inotropic effect in this prepara- tion. The reasons for the opposite effects of quinidine on the ino- tropic action of digoxin, depending on the order of drug addition, are unclear. Other studies in cultured heart cells (Horowitz et al., 1982), guinea pig and rat cardiac muscle (Kim et_al,, 1981a), and ferret papillary muscle (Lash et_al,, 1982), however, have found that quinidine did not influence the rate of onset or the magnitude of the positive inotropic actions of digoxin. In these studies, quinidine had no effect on glycoside inhibition of sodium pump activity (Horowitz et_ al,, 1982; Kim et_a13, 1981a) or on glycoside binding to the Na+,K+- ATPase during incubation of the muscle preparation with digoxin (Kim et al,, 1981a). The quinidine concentrations (3-15 uM) used in these studies were at or near therapeutic antiarrhythmic concentrations and were capable of elevating serum digoxin concentrations in_vivo (Kim et_ 27 al,, 1981c). These studies indicated that quinidine did not diminish the inotropic actions of digoxin in isolated heart muscle preparations. In summary, the question of whether quinidine interacts with the direct effects of digoxin on cardiac muscle function remains unsettled. The hypothesis has been advanced that quinidine may be capable of diminishing the rate of digoxin binding to the Na+,K+-ATPase by de- creasing sodium influx through the sarcolemmal sodium channel. The resulting decrease in intracellular sodium would reduce the glycoside— sensitive conformation of the sodium pump which would decrease the rate of digoxin binding and delay the onset of the positive inotropic effect of digoxin. Experimental studies in man and animals have found, most often, that quinidine does not diminish digoxin effects in the heart. Previous studies in this laboratory (Kim et_a1,, 1981a) have also found that quinidine did not decrease the positive inotropic effect or speci— fic binding of digoxin in isolated cardiac muscle preparations. The reasons for the lack of the expected quinidine/digoxin interaction in many experimental studies are unknown. Thus, the objective of this project was to investigate why quinidine apparently fails to interact with digoxin despite the above hypothesis and to determine if quinidine would reduce the positive inotropic effect of digoxin under certain experimental conditions. The results of this study may help formulate a rationale basis to determine if the serum digoxin concentration in patients receiving combined quinidine/digoxin therapy should be “targeted” to a specified concentration or allowed to rise in the presence of quinidine. MATERIALS AND METHODS A. Inotropic Studies Guinea pigs of either sex weighing 250-450 g were stunned by a sharp blow to the head and the hearts were rapidly removed. After retrograde perfusion of the aorta with a Krebs-Henseleit bicarbonate buffer solution (118 mM NaCl, 4.8 mM KCl, 1.2 mM MgSO 28 mM NaHCO 4, 3’ 1.0 mM KH P0 1.2 mM CaCl2 and 11 mM glucose) to wash out blood, the 2 4’ left atrial muscle was excised and hung vertically in a tissue bath at 32°C containing the above buffer solution saturated with 95% 02-5% C02. Platinum electrodes were used for electrical field stimulation of the atrial muscle with square wave pulses of 4 ms duration and 20% above threshold. Resting tension was adjusted to l g and nearly isometric force of contraction was recorded continuously on a polygraph recorder (Grass Instrument Company, Quincy, MA; Polygraph 7B) equipped with a force-displacement transducer (Grass Instrument Company, FT-e 03C). After an equilibration period of at least 60 minutes, a test drug or vehicle was added to the bathing medium. Quinidine was dissolved in water and benzocaine was dissolved in polethylene glycol. Polyethylene glycol content in the bathing medium did not exceed 0.1%. After the developed tension of the atrial muscle had reached a new steady state, digoxin, in the final concentration of 0.6 uM, was added to the bathing medium. The inotropic effects of digoxin were expressed as the change 28 29 in the force of contraction of the atrial muscle relative to force observed before glycoside administration. The rate of onset of the positive inotropic effect of digoxin was monitored and expressed as the time to half maximal drug effect. The inotropic effects of quinidine or benzocaine alone was also monitored. B. Action Potential Studies Left atrial muscle was isolated from guinea pig heart as described above. The atrial muscle was pinned to the bottom of a constant tempera- ture bath maintained at 31°C and superfused at a rate of 3 ml/min with Krebs-Henseleit bicarbonate buffer solution saturated with 95% 02/5% C02. Muscle preparations were stimulated by an extracellular electrode which delivered square-wave pulses of 4 msec duration, 20% above threshold. Transmembrane potentials were recorded using 3 M KCl-filled micro- electrodes connected to a high input impedance amplifier (WP Instru— ments, New Haven, CT; Model 707) via a Ag/AgCl wire (floating elec- trode). Electrode tip resistances were 10-30 Me. The output of the amplifier was electronically differentiated (Galveston Electronics Corp, Galveston, TX) to measure maximal upstroke velocity of the action potential. The outputs of the amplifier and differentiator were dis— played on an oscilloscope (Tektronix Inc., Beaverton, OR; Model R5103N) and photographed with a kymograph camera (Grass Instruments Corp, Quincy, MA; Model C-3). In some experiments, the amplifier output was also stored and analyzed on a POP-8 computer (Digital Equipment Corp, Maynard, MA). Drug effects on resting potential, maximal upstroke velocity (Vmax) of the action potential, action potential amplitude and 30 duration at 20 (T20), 50 (150) and 90 percent (T90) repolarization were monitored. In the present experiments, continuous impalements were maintained in a single cell throughout the course of the experiment. After stable action potentials were recorded for at least 15 minutes from prepara- tions perfused with a drug—free buffer solution, cumulative dose- response experiments were performed. Data are expressed relative to corresponding values observed during the control period. Where indi- cated, RbCl (5 mM) was substituted for KCl and KH2P04 was replaced with NaH2P04. 86 c. Rb+ Uptake Studies Sodium pump activity was measured in beating left atrial muscle preparations by a modification of the method of Yamamoto et_al, (1979). Muscle preparations were isolated as described above and incubated in a Krebs-Henseleit bicarbonate buffer solution in which the following changes were made: KH2P04 was replaced by NaH2P04 and KCl was replaced with 5 mM RbCl. The RbCl-containing, K+-free Krebs-Henseleit buffer solution (Rb+ K-H) was continuously aerated with 95% 02/5% C02 and maintained at 32°C. The atrial muscle were field-stimulated at the indicated frequency with platinum electrodes which delivered square- wave pulses of 4 msec duration, 50% above threshold from a Grass Instruments S—9 stimulator. After an equilibration period of at least 60 minutes, the test drug was added. After the actions of the drug had 86mm were added to the buffer solution. 86Rb+_ stabilized, tracer amounts of Twenty minutes later, the atria were washed for 1 minute in a free, Rb+ K-H (24°C), blotted, and weighed. Radioactivity in the 31 tissue was quantified using a gamma scintillation spectrometer (Tracor Analytic, Des Plaines, IL; Isocap 300). Ouabain-sensitive Rb+ uptake was determined as the difference in the isotope uptake obtained in the absence and presence of 0.3 mM ouabain. Initial experiments showed 86 that ouabain—sensitive Rb uptake was almost linear up to 30 minutes (Figure 1, inset). 86Rb+ uptake in beating (not sodium—loaded) Ouabain-sensitive heart muscle preparations is thought to be a measure of on-going sodium pump activity because under steady—state conditions, sodium pump acti- vity is in balance with the rate of sodium influx into the cardiac cell (Akera £11., 1981; Eisner std-s l983b). For this reason, 86Rb+ uptake was measured after drug effects had stabilized when steady-state conditions had been reached. Thus, ouabain-sensitive 86Rb+ uptake can also be used as an indirect measure of sodium influx rate in beating atrial muscle preparations. To establish that the experimental condi— tions used could detect charges in sodium influx, the effect of altera- tions in the rate of sodium influx on 86Rb+ uptake was examined. The rate of sodium influx was changed by stimulating the atrial muscle at different frequencies as Na influx rate into heart muscle fibers is reported to be increased by electrical stimulation (Langer, 1974). The results of preliminary experiments show that ouabain—sensitive 86Rb+ uptake was linearly related to frequency of stimulation (Figure 1) which agrees with the data of Yamamoto et_a13 (1979) who also measured 86Rb+ uptake in beating guinea pig left atrial muscle with a 2 mM RbCl Rb+ K-H. Thus, sodium pump activity measured under the conditions outlined above may also serve as a measure of the rate of sodium influx in beating heart muscle. 32 h) tn B) CD nmole/mg/2O min) out a: 8 (fl “no Uptake( l l O 1 2 3 Stimulation rate (Hz) (3 b D b h Figure 1. Effect of stimulation frequency on 86Rb+ uptake. Guinea pig left atrial muscle was incubated in a 5 mM RbCl, K+-free bicarbo- nate buffer solution at 32°C and electrically paced at the indicated frequency. Subsequently, tracer amounts of 5Rb+ was added and uptake allowed to occur for 20 minutes. Inset: 85Rb+ uptake as a function of time was measured in left atrial muscle stimulated at 1.5 Hz. The ordinate is ouabain-sensitive 86Rb+ uptake expressed as nmoles RbT/mg wet weight. Ouabain-sensitive uptake was calculated as the difference in isotope uptake in the absence and presence of 0.3 mM ouabain. Data are expressed as the mean :_S.E.M., n=3. 33 D. [3H10uabain Binding Studies 1. Estimation of digoxin binding to Na+,K+—ATPase in heart muscle preparations: Fractional occupancy The binding of digoxin to the Na+,K+-ATPase in beating left atrial muscle preparations was estimated by homogenizing the atrial muscle and assaying the initial velocity of ATP-dependent binding of [3H]ouabain to the homogenate. After a predetermined period of expo— sure to digoxin at 32°C, the muscle preparation was removed from the Krebs—Henseleit buffer solution and immediately homogenized in 3.5 ml of an ice—cold solution containing 1 mM EDTA and 10 mM Tris-HCl buffer (pH=7.5) using a Dounce ball-type homogenizer. The homogenate was transferred to a Potter-Elvehjem homogenizer, homogenized with a motor- driven Teflon pestle (3000 rpm, 12 sec), and then filtered through a stainless—steel wire mesh (150 uM pore size) to remove tissue debris (Temma and Akera, 1982). The final protein concentration of the homo- genate was approximately 1 mg/ml. An aliquot (0.1 ml) of homogenate was added to 0.9 ml of a prewarmed (37°C) incubation mixture containing 50 nM [3H10uabain, 100 mM NaCl, 5 mM MgCl2 and 50 mM Tris-HCl buffer (pH 7.5) with or without 5 mM Tris-ATP. The binding reaction was terminated after 2 minutes by the addition of 6 m1 of an ice—cold stopping solution containing 15 mM KCl, 0.1 mM unlabelled ouabain and 50 mM Tris—HCl buffer (pH 7.5). This mixture was passed through a nitrocellulose filter (Millipore Corporation, Bedford, MA; type AA, pore size 0.8 um) to separate bound and unbound [3H]ouabain. The filter was then washed with two addi— tional 6 ml aliquots of stopping solution within 10 seconds. The radioactivity trapped on the filter was quantified by liquid scintilla- tion spectrometry after dissolving the filter in 1 ml of ethylene 34 glycol monomethylether (Pierce Chemical Co., Rockford, IL; Pier- solve). The dissociation of bound cardiac glycoside from guinea pig heart Na+,K+-ATPase in ice-cold solution is relatively Slow (Akera 31_ 31,, 1973). Therefore, the release of bound digoxin during the pre— paration of the homogenate is anticipated to be Slight; however, the time between the removal of the atrial muscle from the Krebs-Henseleit buffer solution and the addition of the homogenate to the binding medium was set at 6.5 min to avoid any variability in the amount of digoxin released from tissue binding sites during preparation of the homogenate. ATP-dependent [3H]ouabain binding was calculated by subtract- ing the amount of [3H]ouabain bound in the absence of ATP (nonspecific) from the value observed in its presence. ATP-dependent [3H]ouabain binding represents the specific binding of [3H]ouabain to the glycoside receptor site on the Na+,K+-ATPase (Allen 31_31,, 1971). A reduction in the initial velocity of ATP-dependent binding of [3H]ouabain indi- cates previous occupancy of the glycoside receptor site by digoxin (Ku 31_31,, 1974). 2. [3H]0uabain bindipg to3guinea pig heart microsomes In order to determine the effect of quinidine or benzocaine on the binding of cardiac glycosides to the Na+,K+-ATPase, [3H]ouabain binding to guinea pig heart microsomal Na+,K+-ATPase was assayed in the presence of various concentrations of the drug. Guinea pig heart microsomes were isolated by a modification of the method of Akera 33 31, (1973). Guinea pig cardiac ventricular muscle was minced with scissors and homogenized with 4 volumes of an ice-cold solution II[Z:__________________________________________________________—_——TTT “7 35 containing 0.25 M sucrose, 1 mM NaZEDTA, and 5 mM dl-histidine (pH 7.0) using a Dounce ball-type homogenizer (both loose and tight pestle). The homogenate was transferred to a Potter—Elvehjem homogenizer and briefly (10 sec) homogenized with a motor—driven Teflon pestle (1000 rpm). The final homogenate was centrifuged at 10,000 x g for 15 min at 0°C to remove cell debris and the mitchondrial fraction. The super- natant was then centrifuged at 100,000 x g for 60 min at 0°C. This supernatant was discarded and the pellet was resuspended in an ice-cold solution containing 0.25 M sucrose, 1 mM EDTA, and 5 mM dl-histidine (pH 7.0). This crude microsomal preparation was frozen for later use. [3H]0uabain binding to the microsomal Na+,K+—ATPase was assayed by adding 0.1 ml of the microsomal preparation to 0.8 ml of a prewarmed (37°C) incubation mixture containing 25 nM [3H]ouabain, 20 mM NaCl, 5 mM MgC12, 50 mM Tris-HCl (pH 7.5), and various concentrations of quinidine or benzocaine. In some assays, KCl, at indicated concen- trations, was also present in the incubation mixture. After a further 5 min incubation at 37°C, the glycoside binding reaction was started by the addition of 0.1 m1 of Tris-ATP (pH 6.8) so that the final ATP concentration was 5 mM. Nonspecific [3H]ouabain binding was assayed by the addition of an equal volume of H20, instead of Tris-ATP. The binding reaction was terminated 3 minutes later by the addition of 6 m1 of ice-cold stopping solution (see above). Bound [3H]ouabain was separated by filtration on a nitrocellulose filter and the filters were washed with 2 aliquots (6 ml each) of stopping solution. ATP—dependent [3H]ouabain binding was calculated as the difference in the amount of [3H]ouabain bound in the presence and absence of ATP. Final protein concentration during the binding reaction was 100 ug/ml. 36 3. [3H]0uabain binding to guinea pig heart Na+,K+-ATPase Cardiac Na+,K+—ATPase was obtained from ventricular muscle of guinea pigs by the method of Akera and Brody (1971) as later modified by Ku 31_31, (1976). Binding reactions were carried out essentially the same as those for guinea pig heart microsomes except for the follow- ing modifications: 1) final NaCl concentration was varied between 0- 100 mM with choline chloride being used as an ionic and osmotic substi- tute so that NaCl plus choline chloride equalled 100 mM, 2) KCl was not present in the reaction mixture, 3) the nonspecific binding medium without ATP also contained 0.2 mM unlabelled ouabain, and 4) final protein concentration was 0.5 mg/ml. E. Miscellaneous All chemicals were reagent grade. Digoxin, quinidine hydrochlor- ide and benzocaine (p-ethyl aminobenzoate) were purchased from Sigma Chemial Company (St. Louis, MO). [3H]Ouabain and 86RbCl were purchased from New England Nuclear (Boston, MA) or Amersham Corp. (Arlington Heights, IL). Nadolol was kindly supplied by E.R. Squibb and Sons (Princeton, NJ). Protein concentrations were estimated by the method of Lowry 31_31, (1951) using bovine serum albumin as the standard. Statistical analyses were performed where indicated by use of t- test or one—way or two-way analysis of variance with paired comparisons by least Significant difference. Criterion for significance was a P value less than 0.05. Data are expressed as the mean :_S.E.M. RESULTS A. Effect of Quinidine on the Inotropic Action of Digoxin in Atrial Muscle Preparations of the Guinea Pig Some studies in man (Hirsh 31_31,, 1980; Steiness 31_31,, l980), experimental animal models (Warner 31_31,, 1984), and isolated cardiac muscle preparations (Williams and Mathew, 1981) have indicated that quinidine lessened the effects of digoxin in Cardiac muscle while others have found that quinidine did not influence the effect of digoxin on cardiac mechanical function in man (Zaman 31_31,, 1981; Belz 3§_31,, 1982) and in isolated cardiac muscle preparations (Kim 31_31,, 1981a; Horowitz 31_31,, 1982; Lash 31_31,, 1982). Investigations of the effect of quinidine on the positive inotropic action of digoxin in man and in live animal models have been complicated by changes in serum digoxin concentration. Therefore, the effect of quinidine on the inotropic actions of digoxin was Studied in isolated left atrial muscle of the guinea pig so that cardiac glycoside concentration could be maintained at a constant level. After an initial equilibration period of 60 minutes, the force of contraction of left atrial muscle preparations stimulated at 1.5 Hz remained stable for at least 2 hr. Developed tension after the equili- bration period was 0.78:0.05 9 (mean :_S.E.M. of 21 preparations), and addition of vehicle (H20) had no effect on develOped tension. Quinidine 37 38 produced a positive inotropic effect at 10 and 30 pH (Figure 2). The peak effect on contractile force was reached within 15 minutes of drug addition. With 10 uM quinidine, the increaSe in force was sustained while in the presence of 30 uM quinidine developed tension decreased slightly towards controls after an initial increase and then stabilized by 40 minutes after drug addition. Higher concentrations of quinidine (100 uM; not shown in Figure 2) developed the initial increase in force which, however, was followed by a decline in force that did not stabi- lize. Often, these preparations could not be consistently excited at 1.5 Hz and/or had a substantial increase in the threshold voltage for stimulation. The instability of the atrial muscle preparations in the presence of high drug concentrations made their use in subsequent experiments impossible. The positive inotropic effect of quinidine and its biphasic effects on developed tension observed in the present study are similar to those reported earlier in guinea pig left atrial muscle (Nawrath, 1981). Forty minutes after the addition of 3-30 uM quinidine to the incubation medium, developed tension had stabilized. At this time, digoxin in a final concentration of 0.6 uM was added and the magnitude as well as the rate of onset of positive inotropic effect of the glyco- side was monitored. In atrial muscle preparations which had not been treated with quinidine, 0.6 uM digoxin increased developed tension slowly, reaching a plateau within 30-40 minutes. The time to half maximal inotropic effect, T??2, which reflects the rate of onset of glycoside action, was 1312 min (mean :_S.E.M. of four experiments) and the magnitude of the glycoside—induced increase in contractile force was 0.78:0.13 g. No 39 00 i»; I Bog Chenoe “:1 developed tension (9) ,o h) Figure 2. Effect of quinidine on developed tension of guinea pig left atrial muscle. Muscle preparations were stimulated at 1.5 Hz. After a 60-minute equilibration period, quinidine was added to the incubation medium at the final concentration of 0 (n=4), 3 (n=6), 10 (n=6), and 30 uM (n=5). The peak effect of quinidine on developed tension is expressed as the mean :_S.E.M. Asterisks denote a signi- ficant difference from 0 uM quinidine as determined by 1-way ANOVA. 40 change in resting tension was observed after digoxin administration. Figure 3 shows the positive inotropic effect of 0.6 uM digoxin in the presence of various concentrations of quinidine. The increase in developed tension produced by 0.6 uM digoxin in atrial muscle prepara- tions treated with 3, 10, and 30 uM quinidine was 1.03:0.12, 1.08:0.15 and 0.95:0.12 9 (mean 1_S.E.M. of 5 or 6 experiments), respectively. The glycoside-induced increase in force was not significantly different in the absence or the presence of quinidine, at any concentration tested, as determined by a 1-way analysis of variance. The T?72 for the inotropic action of digoxin in the presence of 3, 10, and 30 uM quinidine was 1412, 13:1 and 12:_minutes, respectively. These values were also not significantly different from the T??2 of digoxin observed in the absence of quinidine. These data indicate that quinidine did not significantly influence the magnitude or the rate of onset of the positive inotropic effect of digoxin in left atrial muscle of the guinea pig. The present results are in agreement with those of Kim and co- workers (l981a) who found that 6 uM quinidine did not affect the posi- tive inotropic actions of digoxin in guinea pig heart Langendorff preparations. The question, then, is why the present experiments and those of others (Kim 31_31,, 1981a; Horowitz 31_31,, 1982; Lash 31_31,, 1982) have not demonstrated an interaction between quinidine and digoxin even though a quinidine-induced reduction in sodium influx is anticipated to decrease digitalis binding. One answer may be that quinidine did not decrease sodium influx into the cardiac muscle fibers under the conditions of the experiments, a point which has not been addressed in most published reports on the quinidine/digoxin 41 24‘ 2.2" 0 ‘0 -: Developed Tension (g) Time (min) Figure 3. Effect of quinidine on the time course of the positive inotropic action of 0.6 uM digoxin. Guinea pig left atrial muscle was stimulated at 1.5 Hz. Forty minutes prior to addition of digoxin (0.6 uM), quinidine in the final concentration of 0 (n=4), 3 (n=6), 10 (n=6), and 30 uM (n=5) was added to the incubation medium. Data are expressed as mean :_S.E.M. 42 interaction. As a result, the objective of the following experiments was to evaluate the effect of quinidine on sodium influx and establish experimental conditions in which quinidine was likely to reduce the rate of sodium influx. B. Effect of Quinidine and Benzocaine on the Transmembrane Action Potential 1. Quinidine Quinidine has been Shown to decrease the maximal upstroke velocity of the cardiac action potential (Johnson and McKinnon, 1957; Vaughan-Williams, 1958). The ionic current flowing during the cardiac action potential upstroke is believed to be carried by sodium ions moving down their electrochemical gradient into the cardiac cell (Weidmann, 1955). For this reason, measurements of the maximal up- stroke velocity have been used by many authors (Chen 31_31,, 1975; Hondeghem and Katzung, 1980; Weld 31 31,, 1982) as an index of the fast sodium current; however, the validity of this relationship has recently been questioned (Strichartz and Cohen, 1978; Cohen 31_31,, 1984). Thus, the measurements of maximal upstroke velocity in this study should be regarded as qualitative indicators of quinidine's effect on sodium current. The action potentials recorded in guinea pig left atrial muscle superfused in drug-free buffer and stimulated at 0.5 Hz had the following parameters (mean :_S.E.M. of eight experiments): resting membrane potential, -78:1 mV; amplitude, 107:2 mV; maximal upstroke velocity, V 180:7 V/sec; duration at the time of 90% repolariza- max’ tion, T90, 6313 msec. The measured values of these action potential 43 parameters were similar to published values (Nawrath, 1981). Single impalements were maintained throughout the entire experiment. Action potential recordings in control experiments remained stable up to 3 hours with a single impalement, although T90 had a tendency to shorten slightly (510%). Figure 4 shows the effects of cumulative doses of quinidine on an atrial muscle preparation stimulated at 0.5 Hz. Quini- dine decreased action potential amplitude in a concentration—dependent manner beginning at 10 uM and increased action potential duration, measured at T90, at a drug concentration as low as 3 uM (data not shown), while 100 uM quinidine increased T by 269142% (mean :_S.E.M. 90 of three experiments) compared to the value observed before drug addi— tion. The effects of quinidine on action potential duration, measured at the time of 20 and 50% repolarization,was prolonged to a similar degree as that measured at T90 (data not shown). These results are consistent with previous reports in guinea pig atrial muscle (Nawrath, 1981), and rabbit (Colatsky, 1982) and canine Purkinje fibers (Mirro 31_31,, 1981). Quinidine also produced a dose-dependent decrease in Vmax (Figure 4, inset). In the experiment shown, Vmax in the absence of quinidine was 171 V/sec. At a concentration of 3 uM, the alkaloid had no effect on Vmax’ but 10, 30, and 100 uM quinidine reduced Vmax to 154, 115, and 45 V/sec, respectively. These concentrations of quinidine were similar to those which have been reported to decrease Vmax in guinea pig atrial (Nawrath, 1981) and ventricular muscle pre- parations (Johnson and McKinnon, 1957; Chen 31_31,, 1975) as well as iNa in rat ventricular myocytes (Lee 31 31,, 1981). Quinidine also increased stimulus latency and threshold so that maintaining regular tissue excitation at high drug concentrations required increasing 44 Figure 4. Effect of quinidine on transmembrane action potential. Guinea pig left atrial muscle was stimulated at 0.5 Hz with square— wave pulses of 4 msec duration, 20% above threshold by an extracellu— lar electrode. At higher quinidine concentrations, it was necessary to increase stimulus voltage to maintain rhythmic excitation. In these cases, stimulus strength was readjusted to 20% above the new threshold. After stable action potentials had been recorded for 15 minutes quinidine concentration was increased to 10, 30, and 100 uM in the superfusing buffer solution every 40 minutes. The bar on the left of the figure marks 0 mV. The horizontal and vertical lines in the lower—left hand corner of the figure show 10 msec and 10 mV calibra— tions, respectively. Inset: Tracings of Vmax- For clarity, the inverted records of Vma are offset from the action potential and displayed on an expanded time scale. The tracings of Vmax from left to right are those recorded in the presence of 0, 10, 30 and 100 uM quindine. The calibration bar next to the Vmax tracings are 50 Volts/ sex (vertical) and 5 msec (horizontal). 45 stimulus strength more than two-fold. Resting membrane potential remained unchanged throughout the experiment except at the highest concentration tested, 300 uM, where a depolarization of a few milli- volts occurred. The onset of the action of quinidine at each concen- tration was slow and the maximum effect was observed only after 30-40 minutes. Superfusion of the tissue preparation in drug-free buffer solution for 60 minutes did not completely reverse the drug effect on the action potential (data not shown). The mean data for the effects of cumulative doses of quinidine on Vmax and T90 for atrial muscle stimulated at 0.5 Hz are Shown in Figures 5A and 5B, respectively. These experiments indicate that in atrial muscle stimulated at 0.5 Hz, quinidine caused a decrease in Vmax and a marked prolongation of the action potential. Reuter and Scholz (1977) have postulated that sodium ions may pass through cardiac sarcolemmal calcium channels and, recently, Falk and Cohen (1983) have shown that sodium pump stimulation during repeti— tive electrical stimulation of canine Purkinje fiber is reduced by D— 600, a calcium channel blocker, and that prolonged membrane repolariza- tions may also stimulate the sodium pump (Falk and Cohen, 1981). Thus, sodium entry during the plateau phase of the cardiac action potential may make a significant contribution to total sodium influx (Langer, 1974). The duration of the atrial muscle action potential is much shorter than that of the cardiac Purkinje fiber. Nevertheless, because quinidine prolongs action potential duration, the drug-induced decrease in sodium influx rate during the upstroke of the action potential might be offset by increased sodium influx during the action potential pla- teau. For this reason, the effects of quinidine on the action potential 46 (hunkflne(ufln Figure 5. Effect of quinidine on atrial muscle action potential at various stimulation frequencies. Atrial muscle preparations were stimulated at 0.5 (n=3), 1.5 (n=5), and 3.0 (n=2) Hz. Cumulative concentrations of quinidine were added to the superfusing buffer solution as described in Figure 2. X, maximal upstroke velocity (top panel), and 19 time to 90% repoTarization (bottom panel), is expressed as the megn percentage of the value before drug addition. The Vma xand T0 of action potentials recorded prior to quinidine addit1on at all0 stimulation frequencies has been normalized to 100%. 47 were examined under conditions where the drug-induced depression of Vmax would be more pronounced. Because quinidine has been shown to produce a frequency-dependent depression of Vmax (Johnson and McKinnon, 1957; Hondeghem and Katzung, 1980), the effects of quinidine were investigated in preparations driven at higher stimulation frequencies. Figure 5A illustrates the effects of quinidine on Vmax of preparations stimulated at 0.5, 1.5, and 3.0 Hz. Quinidine produced a progressively greater decrease of Vmax at higher stimulation frequencies, in agree- ment with earlier studies (Hondeghem and Katzung, 1980). The dose- response curve for quinidine was shifted to the left at faster driving rates so that the drug concentration to cause a 50% decrease in Vmax in preparations stimulated at 3.0 was approximately one-fourth that in atrial muscle stimulated at 0.5 Hz. Figure 5B shows the effects of quinidine on T90 in atrial muscle preparations stimulated at different frequencies. Action poten- tial duration, measured at T90, was 63:3 (n=8), 69:1 (n=10), 65:1 msec (n=6) for preparations stimulated at 0.5, 1.5, and 3.0 Hz, re- spectively. The apparent lack of a marked effect of stimulation fre- quency less than 3.0 Hz on the duration of the atrial muscle action potential was consistent with previous reports (Mendez 31_31,, 1956; Pasmooij 31_31,, 1976). In contrast to the effect of quinidine on upstroke velocity, action potential duration was increased less at higher frequencies. At higher stimulation rates, quinidine produced a greater decrease in Vma and a smaller increase in T90. These charac- x teristics of drug action tend to suggest that quinidine would cause a greater decline in net sodium influx in cardiac tissue paced at higher stimulation frequencies. 48 2. Benzocaine Local anesthetics, such as benzocaine, have been Shown to decrease INa (Schwarz 31 31,, 1977), sodium influx rate (van Zwieten, 1969) and intracellular sodium ion concentration (Deitmer and Ellis, 1980a; Eisner 31_31,, 1983a). Furthermore, benzocaine has been re- ported to slow the onset of the positive inotropic effect of strophan- thidin in sheep cardiac Purkinje fibers (Bhattacharyya and Vassalle, 1981). These results suggested that the local anesthetic may also be capable of reducing glycoside binding by decreasing the rate of sodium influx, similar to the expected action of quinidine. To determine under what conditions benzocaine would reduce the rate of sodium influx into guinea pig atrial muscle, the effects of benzocaine on the action potential of guinea pig left atrial muscle were examined and compared to the effects of quinidine. Benzocaine has been found not to display frequency-dependent depression of Vmax in cardiac Purkinje (Gintant 31 31,, 1983) and INa in ventricular fibers (Sanchez-Chapula 31 31,, 1983). For this reason, the effect of benzocaine on the cardiac action potential was only examined in guinea pig left atrial muscle paced at 0.5 Hz. Super- fusion of the preparations with vehicle, polyethylene glycol, at a final concentration of 0.1%, was found to have no Significant effect on measured action potential parameters. Cumulative doses of benzocaine decreased action potential amplitude and V (Figure 6) in a concen- max tration-related manner in atrial muscle stimulated at 0.5 Hz. The mean data for the effect of benzocaine on Vmax are shown in Figure 7A. A fifty percent decrease in Vmax was calculated to occur at a drug con- centration of 900 uM, ten times that of quinidine to produce a decrease 49 Figure 6. Effect of benzocaine on the transmembrane action poten- tial. Guinea pig left atrial muscle was stimulated at 0.5 Hz with square-wave pulses of 4 msec duration, 20% above threshold with an extracellular electrode. Benzocaine, at concentrations greater or equal to 1000 uM, elevated stimulus threshold. Therefore, stimulus voltage was increased to 20% above the new threshold to maintain rhythmic excitation. After stable action potentials were recorded for 15 minutes, benzocaine concentration was increased in a step-wise manner to the indicated values (final concentration; uM) at 20 minute intervals. The bar on the left of the figure denotes 0 mV. Calibra- tions for 10 msec (horizontal line) and 10 mV (vertical line) are shown in the lower-left hand corner of the figure. Inset: Effect of benzocaine on the rate of depolarization during the action potential. Vmax is displayed offset from the action potential recording and with an expanded time scale for clarity. From left to right the traces of V ax are for 0, 300, 600, 1000, and 2000 uM benzocaine. The spike at tHe beginning of the tracing is due to the end of the stimulus effect. The calibration bar next to the Vmax tracings are 50 Volts/sec (verti- cal) and 5 msec (horizontal). 50 100 fir? A 75* 4 5i so - . °> as - ~ 0 1 {IL L L 1 m f VJ,“ V I' V B HE; ' ° 5 360 560 1000‘ 2000‘ Benzocaine ( uM) Figure 7. Effect of benzocaine on transmembrane action potential. Mean data for three experiments showing the effects of cumulative concentrations of benzocaine on the Vmax (top panel) and action potential duration (bottom panel) of guinea pig left atrial muscle stimulated at 0.5 Hz. V is expressed as a percentage of the value recorded before drug add1 1on. Action potential duration at 20 (T 0), 50 (T50) and 90 (T90)% repolarization is expressed as milliseconds. 51 in Vmax of similar magnitude. These concentrations were Similar to those previously reported to decrease V (Gintant 31_31,, 1983) and max INa (Sanchez-Chapula 31_31,, 1983) in canine cardiac Purkinje fibers and rat single ventricular myocytes, respectively. Like quinidine, the stimulus latency and threshold increased in the presence of benzocaine; however, resting membrane potential did not depolarize at any drug concentration tested (Figure 6). Benzocaine did not produce a signifi- cant increase in action potential duration (Figures 6 and 7B), instead, T90 was slightly decreased or not significantly different than control. Therefore, benzocaine decreased vmax without dramatic changes in action potential duration. 86 C. Effects of Quinidine and Benzocaine on Ouabain-Sensitive Rb+ Uptake in Beating Atrial Muscle Preparations In the beating heart, sodium influx must be balanced by sodium efflux in order to maintain cellular homeostasis under steady-state conditions. Increases or decreases in the rate of sodium influx then, should be matched by an increase or decrease in sodium pump activity, respectively, because the sodium pump is the major mechanism for sodium extrusion in the heart muscle cell. Akera 31_31, (1981) have shown 86 + that under certain conditions, sodium pump activity, measured by Rb uptake, may be an indirect measure of sodium influx rate. Therefore, changes in sodium influx rate caused by quinidine and benzocaine were 86Rb+ estimated by measuring uptake in beating left atrial muscle preparations. 52 1. Quinidine Figure 8 shows the effect of quinidine on ouabain-sensitive 86Rb+ uptake in electrically-driven quinea pig left atrial muscle. In 86Rb+ uptake was initiated 40 minutes after quini- these experiments, dine administration, at a time when the drug effect had stabilized so that the tissue had reached a new steady state. Under these condi— tions, 86Rb+ uptake is not only a measure of sodium pump activity, but an index of sodium influx rate as well. In atrial muscle preparations 86Rb+ uptake was ll.l:0.5 nmol stimulated at 1.5 Hz, ouabain-sensitive Rb+/mg wet weight/20 min (mean :_S.E.M. of 6 experiments) in the ab— sence of quinidine (Figure 8, left panel). Quinidine at 3 and 10 uM 86Rb+ uptake; concentrations did not have a significant effect on however, 30 uM quinidine decreased uptake to 8.110.5 nmole Rb+/mg wet weight/20 min (mean :_S.E.M. of five experiments). Thus, quinidine 86Rb+ uptake only at a concentration of 30 uM in atrial decreased muscle stimulated at 1.5 Hz even though 3 and 10 uM drug concentrations significantly decreased Vmax of action potentials recorded in prepara- tions paced at the same rate. The results of the action potential studies suggest that quinidine would decrease sodium influx more readily in atrial muscle stimulated at higher rates. This conclusion was re-examined by measur- 86Rb+ uptake in preparations stimulated at 3.0 86 ing ouabain-sensitive Hz. In control preprations, Rb+ uptake was 28.6:J.3 nmoles Rb+/mg wet weight/20 min. This value is approximately two and one half times greater than the sodium pump activity observed in control preparations stimulated at 1.5 Hz. An increase in sodium pump activity at a higher 53 15ml, emu 'RbupflkehundquVZMMhu II 8 3" fly. . «n- '1 {'1 5r "" 1 °L' so 0 to an lauhllneufll) Figure 8. Effect of quinidine on ouabain-sensitive 86Rb+ uptake. Guinea pig left atrial muscle were stimulated at 1.5 (left panel) or 3.0 Hz (right panel). Forty minutes after the addition of vehicle 1H 0 or the indicated concentration of quinidine, tracer amounts of 6Rb were added to the 5 mM RbCl, KT-free buffer solution. The 86Rb+ uptake was discontinued after 20 minutes, at which time the tissue was briefly washed in 86Rb+-free buffer solution, weighed and radioacti- vity estimated by gamma scintillation spectrometry. Ouabain-sensitive 85Rb+ uptake was calculated as the difference between uptake values observed in the absence and presence of 0.3 mM ouabain. Data are expressed as the mean + S.E.M. of 5 or 6 experiments. Asterisks (*) denote a significant difference from 0 uM quinidine as determined by 1-way ANOVA. 54 stimulation frequency was expected because faster stimulation rates are believed to increase the rate of sodium influx (Langer, 1974; Yamamoto 8°Rb+ uptake 23:3 and 30:3% at 10 and 31_31,, 1979). Quinidine reduced 20 uM concentrations, respectively, under these conditions. These data appeared to confirm the frequency-dependence of the effect of quinidine 86 Rb+ on sodium influx rate as quinidine produced a greater decrease in uptake at lower drug concentrations in atria stimulated at 3.0 Hz than 1.5 Hz. The preceding experiment demonstrated that the effect of quinidine on sodium influx rate was dependent on both drug concentra- tion and frequency of electrical stimulation. If the working hypothe- sis is correct, quinidine should reduce glycoside binding to the Na+,K+-ATPase in the cardiac cell by decreasing the rate of sodium influx. Conversely, if quinidine does not decrease the rate of sodium influx, digitalis binding should be unaffected. For this reason, it was desirable to establish a set of experimental conditions such that under one condition quinidine would decrease sodium influx rate and under another it would not. The frequency-dependent effect of quini— dine suggested that conditions may be selected such that a given con- 86Rb+ uptake in atrial muscle stimu- 86 centration of drug would decrease lated at a fast rate but have no effect on Rb+ uptake in atria stimu— lated at a slow rate. This point was investigated by observing the 86 effects of 20 uM quinidine on ouabain-sensitive Rb+ uptake in atrial muscle preparations stimulated at 0.5 and 3.0 Hz. Figure 9 Shows the 86Rb+ uptake in atria stimulated at 0.5 effects of 20 pH quinidine on Hz. Included for comparison are the data from the previous experiment showing that 20 uM quinidine significantly decreased sodium pump 55 0.5 Hz 3.0 Hz a + i 1‘ 6 Y ”Rb Uptake (n mole/mg/ZOmin) 0 E11] 1.“... 10 201 O 20 Quinidine (pH) Figure 9. Effect of 20 uM quinidine on ouabain-sensitive 86Rb+ uptake at different stimulation frequencies. Atrial muscle prepara- tions were stimulated at 0.5 or 3.0 Hz either in the absence or presence of 20 uM quinidine. Experiments were performed as described in Figure 8. - Data are expressed as the mean :_S.E.M. of 6 experi- ments. Asterisk denotes a Significant difference from 0 uM quinidine as determined by 2-way ANOVA. 56 activity when preparations were stimulated at 3 Hz. In control pre- parations stimulated at 0.5 Hz, 86Rb+ uptake was 4.9:0.6 nmole/mg wet weight/20 min, several times smaller than sodium pump activity in atrial muscle stimulated at 3 Hz, consistent with the beat-dependent e 86Rb+ uptake in preparations in the presence of influx of sodium. Th 20 pH quinidine was 5.2:0.4 nmole/mg wet weight/20 min, a small but not statistically significant increase (Figure 9). These data confirm that a single concentration of quinidine could have different effects on sodium pump activity depending on the frequency of electrical stimula— tion. Catecholamine release has been reported to stimulate uptake in canine cardiac ventricular tissue (Hougen 31 31,, 1981). Sodium pump activity was measured in atrial muscle stimulated with field electrodes using large currents. Under these conditions, both muscle and nerve endings are stimulated in the atrial tissue. The local anesthetic actions of quinidine might antagonize catecholamine release from the nerve endings, similar to the effects of tetrodotoxin in field-stimulated atria (Katz and Kopin, 1969). The apparent de- 86Rb+ uptake caused by quinidine might, in part, or complete- ly be due to inhibition of catecholamine-induced stimulation of 86Rb+ crease in uptake rather than a reduction in sodium influx into muscle fibers. In order to examine this possibility, left atrial muscle preparations stimulated at 3 Hz were incubated in a Krebs—Henseleit buffer contain- ing 1 uM nadolol, a B-adrenoceptor antagonist, beginning 20 minutes before quinidine (20 uM)_addition. This concentration of nadolol was sufficient to shift the concentration of isoproterenol which produced a half-maximal inotropic effect in atrial muscle preparations from 2 to 57 150 nM (data not shown). In preparations treated with nadolol only, 86Rb+ uptake was 25.911.1 nmole Rb+/mg wet weight/20 min (data not 86 shown) which was slightly but not significantly less than Rb+ uptake in untreated atrial muscle (28.6jfl.3 nmol Rb+/mg wet weight/20 86Rb+ min). In nadolol-treated preparations, 20 uM quinidine reduced uptake by 25:3% (data not shown), similar to the degree of reduction of active monovalent cation transport seen in the absence of s-adrenoceptor blockade (30:3%). These data suggest that most, if not all, of the reduction in 86 Rb+ uptake induced by quinidine can be attributed to a drug effect directly in the muscle fibers. 2. Benzocaine The effect of benzocaine on 86 Rb+ uptake was also examined in atrial muscle preparations to establish the experimental conditions in which the local anesthetic agent decreased sodium influx rate. Because the reduction of action potential V by benzocaine is not beat- 86 max + . Rb uptake were measured 1n dependent, the effects of benzocaine on preparations stimulated at 3 Hz only. Actionpotential experiments and preliminary experiments which examined the inotropic actions of benzo— caine showed that drug effects had reached a plateau within 15 minutes. In these experiments, 86Rb+ uptake was begun 20 minutes after benzo- caine administration, when the atrial muscle preparations had reached a new steady state. Figure 10 shows the effects of 20 and 300 uM benzo- 86Rb+ caine on ouabain-sensitive uptake. The low concentration of benzocaine (20 uM) had no effect on 86Rb+ uptake when compared to control values. Benzocaine at a concentration of 300 uM, which de- creased action potential Vmax approximately 15%, caused a 32:11% reduc— 86 tion in Rb+ uptake. The highest concentration of this drug tested, 1 58 Ami 1 C 325. . Eml- . T g... . 5 gflp c1 i;:151r ‘ . 4 ii: _Cl (J 2!) 13C!) Benzocaineofl) Figure 10. Effect of benzocaine on ouabain-sensitive 86Rb+ uptake. Guinea pig left atrial muscle was stimulated at 3.0 Hz. Benzocaine, at the indicated concentration, or vehicle (polyethylene gl§col was added to the incubation medium 20 minutes before beginning 6Rb uptake. Data is expressed as the mean 1 S.E.M. of 3 experiments. Asterisks denote a significant difference in 86Rb+ uptake from 0 uM benzocaine as determined by l-way ANOVA. 59 8°Rb+ uptake by 47:12%, but mM (not shown in Figure 10), decreased visual inspection of the preparations revealed that they were not rhythmically contracting at 3 Hz. Thus, high concentrations of benzo- caine reduced sodium pump activity. In summary, both quinidine and benzocaine reduced vmax of the action potential of guinea pig left atrial muscle fibers. The magni- tude of the reduction of vmax by quinidine was dependent on the stimu- lation rate whereas that of benzocaine has been reported to be inde- pendent of frequency at stimulation rates as fast as 4 Hz (Gintant 33 31,, 1983). Quinidine also caused a dose-dependent prolongation of the action potential which was more pronounced at lower- stimulation fre- quencies in contrast to benzocaine which did not significantly increase action potential duration at the concentrations and stimulation fre- quency tested. Both drugs decreased sodium pump activity in beating muscle preparations. Quinidine produced a dose and frequency-dependent 86Rb+ uptake, similar to its frequency- 86 decrease in ouabain-sensitive dependent effect on Vmax' Benzocaine inhibited Rb+ uptake at con- centrations an order of magnitude higher than quinidine. Similar relative potencies of the two agents were noted during action potential studies. It seems reasonable, therefore, to examine l) the effects of quinidine on the positive inotropic action of digoxin in atrial muscle preparations stimulated at higher frequencies and 2) the effects of high concentrations of benzocaine on the rate of onset and the magni- tude of the positive inotropic action of digoxin. 60 D. Inotropic Action of Digoxin in the Presence of Quinidine or Benzocaine 1. Quinidine Since the rate of onset of glycoside action varies with drug concentration (Park and Vincenzi, 1975), a concentration of digoxin was chosen so that, at the different stimulation rates examined, a substan— tial inotropic effect could be observed without the appearance of toxicity. In preliminary experiments, this concentration was found to be 0.6 uM. Previous experiments demonstrated that the action of quini- dine on both action potential Vmax and sodium pump activity was fre— quency—dependent. Quinidine at a concentration of 20 uM, reduced sodium pump activity in atrial muscle stimulated at 3.0 Hz but not at 0.5 Hz. The goal of these experiments was to determine the effects of quinidine on the positive inotropic effect of cardiac glycosides under conditions in which the antiarrhythmic agent decreased the rate of sodium influx and in those which it did not decrease the rate of sodium influx. For this reason, the inotropic effects of digoxin were in- vestigated in the presence and absence of 20 uM quinidine in left atrial muscle preparations of the guinea pig heart stimulated at 0.5 or 3.0 Hz. At the end of a 60 minute equilibration, atrial muscle pre- parations stimulated at 0.5 or 3.0 Hz produced a 0.26:0.02 9 (mean : S.E.M. of 14 experiments) and 1.09:0.07 9 (mean i_S.E.M. of 22 experi- ments) developed tension, respectively. The atrial muscle preparations stimulated either at 0.5 or 3.0 Hz maintained stable force for at least 2 hours after the equilibration period. Quinidine produced a small positive inotropic effect in preparations which were stimulated at 61 either 0.5 or 3.0 Hz. In atrial muscle paced at 0.5 Hz, the inotropic effect developed slowly, reaching a plateau in approximately 30 minutes, at which time force was increased by 29%. The inotropic effect of quinidine in atrial muscle stimulated at 3.0 Hz developed rapidly, reaching a peak within 2 minutes. Thereafter, contractile force declined towards control levels and plateaued at a level not significantly different from control force by 30 minutes after quini- dine addition. These results are similar to those observed under 1.5 Hz stimulation (Figure 2). Digoxin at a final concentration of 0.6 uM was added to the preparations only after force had stabilized. In the absence of quinidine, the positive inotropic action of digoxin developed slowly and reached a plateau by approximately 100 minutes in preparations stimulated at 0.5 Hz (Figure 11). The T172 for the inotropic action of digoxin in these experiments was 4014 minutes (Figure 12) and the magnitude of the positive inotropic effect was 1.16:0.13 9 (mean :_S.E.M. of 7experiments). The positive ino— tropic effect of digoxin in atrial mascle stimulated at 3.0 Hz de— veloped much more rapidly than that seen in preparations paced at 0.5 Hz. The T?72 was 811 min and the maximal change in developed tension was 0.89:0.07 9 (mean :S.E.M. of 11 experiments). A 2-way analysis of variance revealed that the magnitude of the positive inotropic effect of digoxin, expressed as change in grams developed tension, was not significantly different between preparations stimUlated at 0.5 and 3.0 Hz; however, the rate of onset of the inotropic effect was Signifi- cantly faster in atrial muscle stimulated at 3.0 Hz. The frequency- dependent onset of cardiac glycoside action is similar to that noted by 62 13 12. 05*: 30H: , 1A» 0 A10. , 2. 20 .§o9» o , g 0.8 ,. .. £0]. . § 0.5 . 20 g 0.5 - M(w) mumwu). éoa. . 5103» a Q2» 4 on. 2’ . <1 {6 it 56 do do do in do do {i113 1b 26 :5 do so ‘flnn(mfl0 Figure 11. Effects of quinidine on the time course of the inotropic actions of digoxin in atrial muscle preparations stimulated at 0.5 and 3.0 Hz. Guinea pig left atrial muscle was stimulated at 0.5 (n=7) or 3.0 (n=ll). After a 60-minute equilibration period, quinidine (20 uM) or vehicle (0 uM) was added to the incubation medium. After an addi- tional 40 minutes, digoxin in a final concentration of 0.6 uM was administered (time zero). Data are expressed as mean :_S.E.M. 63 0.5 Hz 3.0 H1 '5‘: $3 H~ + p O 1 Change in developed imam 8 .31.. j: 83 15 13 c» i: i: E L 1 i 0"“0 ,. + «1 LL... __1, DDJ 0 O 20 20 Chunkflne(ul0 ‘I2 8 V 09 Figure 12. Effect of 20 uM quinidine on the positive inotropic effect of digoxin at different stimulation frequencies. Guinea pig left atrial muscle was stimulated at 0.5 or 3.0 Hz. Forty minutes after addition of 0 or 20 uM quinidine, digoxin (0.6 uM) was added to the incubation solution. The maximal change in developed tension (top panel) and the T972, time to half-maximal effect (bottom panel), are shown as the meah :_S.E.M. for 6 experiments at 0.5 Hz and 11 ex- periments at 3.0 Hz. 64 others (Park and Vincenzi, 1975; Bentfeld 31_31,, 1977; Temma and Akera, 1982). The positive inotropic effect of 0.6 uM digoxin was examined in the presence of 20 uM quinidine. In atrial muscle stimulated at 0.5 Hz, in which the quinidine was found to have little effect on the rate of sodium influx, the magnitude of the positive inotropic effect of digoxin was not significantly different in the presence or absence of 20 pH quinidine (Figure 12). The rate of onset of glycoside action was more rapid in the presence of quinidine; however, the difference from the rate of onset in the absence of quinidine was not significant as determined by 2-way ANOVA. In preparations paced at 3.0 Hz, the magnitude of the glycoside-induced increase in developed tension in the presence of 20 uM quinidine was slightly, but not Significantly, smaller than in the absence of quinidine. The rate of onset of digi- talis action was also not significantly influenced by quinidine (Figure 12). The lack of an effect of quinidine on the positive inotropic action of digoxin in preparations stimulated at 0.5 and 3.0 Hz, con- firmed the experimental results observed in atrial muscle stimulated at 1.5 Hz. These experiments, therefore, indicated that quinidine did not influence the positive inotropic effect of digoxin even in experi— mental conditions where quinidine apparently reduced the rate of sodium influx into cardiac muscle cells. 2. Benzocaine Quinidine did not Slow the development of the positive ino- tr0pic effect of digoxin under conditions where the alkaloid appeared to reduce sodium influx and, therefore, was expected to influence glycoside action. Benzocaine, like quinidine, was shown to be able to 65 decrease the rate of sodium influx in beating cardiac muscle. Thus, effects of benzocaine on the inotropic actions of digoxin were examined under experimental conditions where the local anesthetic agent de- creased the rate of sodium influx. Atrial muscle preparations were stimulated at 3.0 Hz. Benzocaine, at a concentration of 20 uM, had no significant effect on the developed tension of the preparations, while a concentration of 300 uM produced a negative inotropic effect which plateaued within 15 minutes, at which time, developed tension had decreased by 28:5%. Digoxin was added to the bathing medium 20 minutes after benzocaine administration. In the presence of 20 and 300 uM benzocaine, the magnitude of the positive inotropic effect of 0.6 uM digoxin was slightly, though not significantly,greater than in the absence of the local anesthetic. The T??2 of the glycoside, though, was significantly slowed by 300 uM benzocaine (Figure 13), a concen- tration of drug Shown to decrease the rate of sodium influx into beating atrial muscle fibers. These experiments demonstrated that the rate of onset of digoxin's positive inotropic effect could be delayed by benzocaine under conditions in which benzocaine decreaSed sodium influx rate. E. Digoxin Binding in Beating Heart Muscle Preparations The hypothesis that quinidine interacts with positive inotropic actions of digoxin is based upon the premise that quinidine would decrease the amount of sodium available to the sodium pump, and that rate of glycoside binding is determined by the amount of sodium. Therefore, the effect of quinidine on digitalis binding to the sodium pump was estimated in beating atrial muscle preparations. 66 3 E to, . (313. "1f'r 1 Eco. FF' . (L44- 4 £5 Q2, a iZIDP ., 1 161+ p-f-T F1i-i E ‘2’ p17 ‘ is: .. - 4.. ct ()L d.__-. A—-—1 ih-J .i 0 20 300 Benzocaine(pM) Figure 13. Effect of benzocaine on the positive inotropic action of digoxin. Guinea pig left atrial muscle was stimulated at 3.0 Hz. After 20 minutes in the presence of 0 (n=10), 20 (n=7) and 300 uM (n=10) benzocaine, digoxin (0.6 uM) was added to the bathing medium. Data for the maximal change in developed tension (t0p panel) and time to half-maximal effect (bottom panel) are expressed as mean :_S.E.M. The asterisk indicates a significant change from 0 uM benzocaine by 1- way ANOVA. 67 l. Quinidine Digoxin binding to the Na+,K+-ATPase in the atrial muscle was examined by determining the fractional occupancy of glycoside receptor sites in preparations incubated with digoxin in the presence and ab— sence of quinidine. Fractional occupancy was estimated from a reduc— tion in the initial velocity of [3H]ouabain binding reaction to homo- genates of the atrial muscle. The initial velocity of ATP-dependent [3H]ouabain is proportional to the number of unoccupied glycoside receptor sites in the homogenate which, in turn, is determined by the number of Na+,K+-ATPase enzyme molecules to which digoxin has bound during the prior incubation with the beating atrial muscle. Thus, digoxin binding to the heart muscle, the fractional occupancy, is inversely proportional to the initial velocity of [3H]ouabain binding, i.e. a decrease in initial velocity corresponds to increased digoxin binding in the beating heart. To determine if quinidine altered the rate of digoxin bind- ing, fractional occupancy was examined at the time of the half maximal inotropic effect of digoxin. Previous experiments showed that the T??2 for digoxin in atrial muscle stimulated at 3 Hz was eight minutes. For this reason, fractional occupancy in atrial muscle driven at 3.0 Hz was estimated after an eight-minute incubation with digoxin. In these experiments, quinidine in a final concentration of 20 uM was added to the incubation medium 40 minutes prior to glycoside addition. Digoxin or vehicle was then added to the buffer solution and after an addi- tional 8 minutes, the preparation was homogenized to measure fractional occupancy. The initial velocity of ATP-dependent binding to homoge- nates of preparations exposed to 20 uM quinidine only was not 68 significantly different from the initial velocity of binding in control preparations exposed to neither quinidine or digoxin. Table 1A shows that the initial velocity of [3H]ouabain binding in homogenates ob- tained from atrial muscle after an 8 minute (T$72) incubation with 0.6 uM digoxin in the absence of quinidine was also not significantly different from control preparations. In contraSt, the initial velocity of [3H]ouabain binding in homogenates of preparations incubated with 0.6 uM digoxin and 20 uM quinidine was significantly less than that of atrial muscle incubated with 20 pH quinidine only or with 0.6 uM digoxin in the absence of quinidine. These observations did not support the hypothesis that quinidine would decrease digoxin binding to Na+,K+-ATPase in atrial muscle under conditions where quinidine de- creased the rate of sodium influx. For this reason, digoxin binding was estimated again under the same experimental conditions except that the digoxin concentration was increased to 2.0 pM. Tissue homogenates of preparations incubated in the presence of 2 uM digoxin for 8 minutes showed a Significant decrease in initial velocity of [3H]ouabain bind- ing; however, in contrast to the previous experiment, binding was not influenced by 20 pM quinidine. Digoxin binding was also estimated at the time of its maximal inotropic effect in atrial muscle stimulated at 3.0 Hz. In these experiments, tissue preparations were incubated with 0.6 uM digoxin for 45 minutes in the presence or absence of 20 uM quinidine. The initial velocity of[3H]ouabain binding was signifi- cantly reduced in atria incubated with 0.6 uM digoxin. The presence or absence of 20 uM quinidine, however, did not influence digoxin binding (Table 1B). Taken together, these experiments indicate that 20 uM quinidine did not change digoxin binding to beating left atrial muscle. 69 TABLE 1 Effect of Quinidine on Fractional Occupancy by Digoxin in Beating Left Atrial Muscle Quinidine (uM) Digoxin (uM) O 20 ATP-dependent [3H]ouabain binding (fmole/mg prot./2 min) A. Incubation Time: 8 minutes 0 555110 551 :25 0.6 582124 490:13*# 2.0 ' 457:24* 449118" B. Incubation Time: 45 minutes 593:14 604147 0.6 51o:13* 491:24* Fractional occupancy of glycoside binding sites by digoxin was3esti- mated from the reduction in the initial velocity of ATP-dependent [ H]- 0uabain binding to homogenates of guinea pig left atrial muscle. Guinea pig left atrial muscle, stimulated at 3 Hz, was incubated with or without 20 uM quinidine for 40 minutes prior to digoxin administration. After the indicated time in the presence of digoxin or vehicle (EtOH), the atrial muscle was homogenized and the initial velocity of ATP-dependent [3H]ouabain binding was assayed as described in the Methods section. A significant difference between digoxin binding, as estimated by frac— tional occupancy, in atrial muscle incubated in the presence and absence of the indicated concentration of digoxin is designated by an asterisk (*) and a significant difference in digoxin fractional occupancy at a given digoxin concentration in the presence and absence of 20 uM quini- dine is indicated by the cross-hatch (#). Statistical significance was determined by 2-way ANOVA. Data are expressed as mean :_S.E.M. of five experiments. ’ 70 2. Benzocaine Inotropic studies showed that benzocaine delayed the onset of the positive inotropic effect of digoxin. To determine if the decrease in the rate of onset was due to slower glycoside binding, the frac- tional occupancy of digoxin in atrial muscle stimulated at 3.0 Hz was estimated at the T??2 of digoxin (Table 2). No Significant reduction in the initial velocity of ATP-dependent [3H]ouabain binding was ob— served in the homogenates of atrial muscle eXposed to 300 uM benzocaine only. The initial velocity of [3H]ouabain binding to homogenates of tissue preparations incubated for 11 min (T???) with 0.6 uM digoxin (data not shown in Table 2) was not significantly different from that in vehicle controls; however, initial velocity of [3H]ouabain binding in homogenates of preparations incubated with 2.0 uM digoxin for 11 min was significantly decreased. The fractional occupancy of the Na+,K+- ATPase by digoxin, however, was not significantly changed by 300 uM benzocaine. These data indicate that the benzocaine-induced delay in the onset of the positive inotropic effect of digoxin might not be related to a decrease in the rate of glycoside binding. In summary, quinidine caused a frequency and dose-dependent decline in both maximal upstroke velocity of the action potential and sodium pump activity which suggested that sodium influx into the cardiac cells was decreased by quinidine. A decrease in sodium influx rate has been postulated to diminish the rate of cardiac glycoside binding to the sodium pump. Quinidine would then be expected to delay the development of the positive inotropic effect of cardiac glycosides under conditions where the agent decreased sodium influx rate. 71 TABLE 2 Effect of Benzocaine on Digoxin Fractional Occupancy in Guinea Pig Left Atrial Muscle Benzocaine (uM) Digoxin (uM) O . 300 ATP-dependent [3H]ouabain binding (fmole/mg prot./2 min) 0 531 :30 573124 2.0 443:31* 454:27* Atrial muscle preparations stimulated at 3 Hz were incubated in the absence or presence of 300 uM benzocaine for 20 minutes before addition of digoxin. The atrial muscle was homogenized after an additional 11 minute incubation in the presence of 2.0 uM digoxin and fractional occu- pancy of the receptor site by digoxin was assayed as described in METHODS. A significant difference in the fractional occupancy between atrial muscle incubated in the 0 and 2.0 uM digoxin as determined by 2-way ANOVA is indicated by an asterisk. Data are expressed as the mean :_S.E.M. of six experiments. 72 Experimental results under these conditions showed that quinidine did not slow the development of the inotropic actions of digoxin nor de- crease binding of digoxin to the Na+,K+-ATPase in beating atrial muscle preparations. Benzocaine also was shown to decrease action potential Vmax and sodium pump activity but, unlike quinidine, the rate of onset of the positive inotropic action of digoxin was diminished by benzo— caine under conditions where the local anesthetic agent decreased the rate of sodium influx. The rate of digoxin binding to the Na+,K+— ATPase in intact preparations, however, was not deCreased by benzo- caine. Thus, two drugs, both of which appeared to decrease the rate of sodium influx, did not influence the binding of digoxin to the sarco- lemmal Na+,K+-ATPase in cardiac muscle fibers. These data do not support the working hypothesis that glycoside binding to the sodium pump is dependent on the level of intracellular sodium in the heart cell. The following experiments will examine possible reasons for the apparent discrepancy. 3. Effect of quinidine on the digoxin binding to beating heart muscle incubated in rubidium-containinggbuffer solutions The working hypothesis is based on the concept that digoxin binding is dependent on the rate of sodium influx into cardiac muscle fibers, yet the data indicate that quinidine and benzocaine could decrease sodium influx rate without changing digoxin binding to the Na+,K+-ATPase in beating atrial muscle. This apparent discrepancy between the working hypothesis and the experimental data might be explained as an artifact of the different experimental procedures used to measure sodium influx rate and digoxin binding. The rate of sodium f 86 influx was estimated by the rate 0 Rb+ uptake in beating atrial muscle. In these experiments, the cardiac tissue was incubated in a 5 73 mM RbCl, K+-free bicarbonate buffer solution, whereas for the measure- ment of digoxin binding, muscle preparations were incubated in a 5.8 mM K+, Rb+-free bicarbonate buffer. Rubidium was used as an ionic sub- stitute for potassium-activation of the Na+,K+-ATPase; however, Rb+ has a Slightly greater potency than K+ for dephosphorylating the potassium-sensitive conformation of the Na+,K+-ATPase and the enzyme/ cation complex formed after dephosphorylation is more stable in the presence of Rb+ than in the presence of K+ (Post 31 31,, l972).’ As a result, availability of the glycoside-sensitive conformation of the Na+,K+-ATPase would be less in the presence of Rb+ than in the presence of potassium. Rubidium passes through sarcolemmal potassium channels; however, experiments in resting frog sartorius muscle (Sjodin, 1959) and squid axon (Hagiwara 31_31,, 1972) suggest that membrane permeabi- lity for Rb+ is less than that for potassium. The membrane permeabi— lity for Rb+ might affect fluxes of other ions across the sarcolemma. These differences between Rb+ and K+ ions raised the possibility that 86Rb+ uptake in beating atrial muscle might the effects of quinidine on not be entirely applicable to the effect of the alkaloid on sodium pump activity under the experimental conditions used to measure digoxin binding. For this reason, digoxin binding was measured in atrial muscle incubated in a Rb+-containing buffer solution. Quinidine (20 uM) significantly decreased 86Rb+ uptake in atrial muscle stimulated at 3 Hz. Therefore, digoxin binding was measured in atrial muscle preparations incubated in a modified K-H bicarbonate buffer solution containing 5 mM RbCl and stimulated at 3 Hz. Preliminary experiments demonstrated that 20 pM quinidine did not significantly alter the fractional occupancy of glycoside binding sites 74 by digoxin in atrial muscle which had been incubated for 11 minutes with 0.5 or 2.0 uM digoxin. The initial velocity of [3H1ouabain bind- ing was Significantly decreased in homogenates of atrial muscle exposed to 0.6 uM digoxin for 45 minutes (Table 3). Digoxin binding in those preparations incubated with digoxin and 20 uM quinidine was not signi- ficantly different from glycoside binding in preparations exposed to digoxin alone. These data demonstrate that under the same conditions 86Rb+ uptake, digoxin binding in where quinidine was shown to decrease left atrial muscle was not influenced by the presence of quinidine. The discrepancy between the working hypothesis and the previous data, therefore, cannot be explained as an artifact of the different methods used to measure sodium pump activity and digoxin binding. F. Effect of Quinidine on [3H]Ouabain Binding to Na+,K+-ATPase l. [3H]Ouabain binding to guinea pig heart microsomes The data to this point are inconsistent with the hypothesis that digoxin binding to the glycoside-sensitive form of the Na+,K+- ATPase in the intact cell is dependent on the rate of sodium extrusion by the sodium pump. The relationship between the rate of sodium ex- trusion by the sodium pump, the availability of the glycoside-sensitive conformation of the Na+,K+-ATPase and glycoside binding is complex and could be modified by changes in the effective affinity of the sodium pump for the glycoside. The data showing that quinidine caused a decrease in sodium pump activity with no change in digoxin binding could suggest that quinidine stimulated binding of the glycoside to the Na+,K+-ATPase. This could occur if quinidine stabilizes the glycoside— sensitive form of Na+,K+-ATPase thereby stimulating digoxin binding to 75 TABLE 3 Effect of Quinidine on Digoxin Fractional Occupancy in Atrial Muscle Incubated in Rb+-Containing Buffer solution Quinidine (uM) Digoxin (uM) O 20 ATP-dependent [3H]ouabain binding (fmole/mg prot./2 min) 0 407:16 420_+_21 0. 6 299:49* 259:19* Left atrial muscle preparations of guinea-pig heart were bathed in a 5 mM RbCl, K+—free buffer solution and electrically paced at 3 Hz. Fractional occupancy of glycoside receptor sites by digoxin was esti- mated in atrial muscle incubated with O or 0.6 uM digoxin for 45 minutes. Quinidine or vehicle (H20) was added to the bathing solution 40 minutes prior to digoxin. An asterisk indicates a significant difference in fractional occupancy between atrial muscle exposed O and 0.6 uM digoxin as determined by 2-way ANOVA. Data are expressed as the mean :_S.E.M. of four experiments. 76 Na+,K+-ATPase. In order to investigate this possibility, the binding of [3H]ouabain to guinea pig heart microsomes was measured in the presence of various concentrations of quinidine. Quinidine could stimulate binding to the Na+,K+-ATPase by increasing the apparent affinity (decreasing the apparent KB) of digoxin for the enzyme. A decrease in apparent KD could occur by decreasing the dissociation rate constant or increasing the apparent association rate constant. Quinidine, even in high concentrations, has been shown to have no effect on the rate of dissociation of digoxin from lamb kidney Na+,K+-ATPase (Ball 31 31,, 1981). For this reason, the effect of quinidine on the rate of association of [3H]ouabain with guinea pig heart Na+,K+-ATPase was investigated. The time course of ATP-dependent binding of 25 nM [3H]ouabain in the presence of 20 nM NaCl, 5 mM MgC12, and 5 mM ATP was almost linear for up to 3 minutes (Figure 14, inset). Therefore, 3 minute binding times were used in subsequent experiments with cardiac microsomes as an approximation of the initial rate of ouabain binding to the Na+,K+-ATPase. Sodium concentration in these experiments was 20 mM, a concentration which submaximally stimulates ouabain binding. This low concentration was used to ensure that a quinidine-induced stimulation of binding could be observed and because the sodium pump is most likely to be activated by submaximal concentration of Na+ in beating heart muscle preparations. In the absence of quinidine, the initial velocity of [3H]ouabain bind- ing to the Na+,K+-ATPase in these experiments was 784119 fmole/mg prot./3 minutes (Mean :_S.E.M. of three experiments). Quinidine did not stimulate ouabain binding to the Na+,K+-ATPase at any concentration § Li—v/L ‘ ‘ ‘ ‘ 0 0.1 0.3 1.0 3.0 10.0 KC! (mu) [woo-hem bound (t mole/mo W3mh) C) Figure 14. Effect of K+ on ATP-dependent [3H]ouabain binding to cardiac microsomes. Guinea pig heart microsomal preparations were preincubated for 5 minutes at 37° C in the presence of various concen- trations of quinidine with 50 mM Tris- HCl (pH 7. 5), 20 mM NaC1,5 mM MgC12, and 25 nM [3 H]0uabain with or without 0. 6 mM KCl. The binding reaction was started by the addition of 5 mM Tris-ATP (pH 6.8) or an equal volume of H20. Three minutes later the binding reaction was stopped by the addition of an excess volume of a solution containing 15 mM KCl, 0.1 mM ouabain and 50 mM Tris-HCl buffer (pH 7.5) and the m1xture was immediately filtered to separate bound from unbound [ H]0uabain. ATP-dependent binding was calculated as the difference between the amount of radioactivity bound in the presence and absence of ATP. Values are the mean of 2 experiments. Final protein concen- tration was 100 ug/ml. Inset: Time-course of [3H]ouabain binding. Binding assays were performed as above in the absence of K except that the binding reaction was stopped at the indicated times. Verti- cal axis: ATP- -dependent [3 H]0uabain bound (fmole/mg prot. /3 minutes). 78 tested. Instead quinidine caused a dose-dependent decrease in the initial velocity of [3H]ouabain binding in the presence of 20 nM NaCl, 5 mM MgCl2 and 5 mM ATP (Figure 15). The drug concentration to produce 50% inhibition of binding, however, was approximately 600 uM. Inhibi- tion of binding was minimal at concentrations of quinidine shown to inhibit sodium pump activity in beating left atrial muscle prepara- tions. The effect of high concentrations of quinidine on digoxin binding has been reported to be ligand-dependent (Ball 31_31,, 1981). 2+ and ATP favored The above condition for binding reaction with Na+, Mg the formation of the phosphorylated intermediates of the Na+,K+- ATPase; however, in the beating heart muscle, K+ ions are available and are most likely to act on the phosphorylated enzyme to promote dephosphorylation (Post 31_31,, 1965; Sen 31_31,, 1969) and to reduce the glycoside-sensitive conformation of the Na+,K+-ATPase (Matsui and Schwartz, 1968; Akera and Brody, 1971). Quinidine may antagonize the actions of K+ on the Na+,K+-ATPase, so the effect of quinidine on 2+, ATP, and [3H]ouabain binding was examined in the presence of Na+, Mg K+. Since the dephosphorylated forms of the Na+,K+-ATPase do not readily bind cardiac glycosides (Matsui and Schwartz, 1968), it was first necessary to find a concentration of K+ which only partially inhibits [3H]ouabain binding. Figure 14 shows that K+ caused a con- centration-dependent decrease in ATP-dependent ouabain binding. Half- maximal inhibition of binding occurred at 0.6 mM KCl, a concentration which was used in the following experiments. The rate of [3H]ouabain 2+ binding measured in the presence of Na+, Mg , ATP, and K+ was 387118 fmoles/mg prot./3 minutes (Mean :_S.E.M. of three experiments). A 79 pang) § § [fiflouabeuibound(fnx*mhnu 8 8 is do 160 300 1000 WNW"), ATP- C’ 0’ a» Figure 15. Effect of quinidine on ATP—dependent [3H]ouabain binding to cardiac microsomes. Binding assays were performed as described in Figure 14, in the presence or absence of 0.6 mM KCl, except that various concenrations of quinidine were included. Data are expressed as mean :_S.E.M. of 3 experiments. 80 dose-dependent decrease in [3H]ouabain binding was produced at high concentrations of quinidine. A slight though not statistically signi- ficant, stimulation of glycoside binding was noted at 3 and 10 uM drug concentration; however, the quinidine-induced stimulation of glycoside binding, if it occurs, seems to be too small to explain the discrepancy between the effects of quinidine on the rate of sodium influx and the lack of effect of the alkaloid on digoxin binding in beating cardiac muscle preparations. 2. [3H]Ouabain Binding to Partially Purified Guinea Pig Heart Na+,K+—ATPase in the Presence of Low Sodium ConCentrations Sodium has been Shown to stimulate the rate of cardiac glyco- side binding to the Na+,K+—ATPase, but in these studies, the effects of sodium were examined at concentrations greater than the Km for sodium binding to the enzyme (13.7 mM; Lindenmayer and Schwartz, 1973). Ouabain binding at nonsaturating concentrations (510 mM), however, is greater than predicted by an enzyme model assuming a single sodium binding Site. Instead, a model with a second, high affinity sodium binding site on the Na+,K+-ATPase with a Km for sodium of less than 1' mM has been developed to explain the unexpectedly high glycoside bind- ing at low sodium concentrations (Inagaki 31_31,, 1974). These data point out that sodium stimulation of ouabain binding at low sodium concentrations, similar to those observed in cardiac muscle fibers with ion-sensitive electrodes (Cohen 31.31,, 1982), cannot be extrapolated from the effect of high sodium concentrations of ouabain binding. Quinidine is expected to decrease intracellular sodium ion concentra— tion, but the magnitude of this decrease is unknown as the effect of quinidine on intracellular sodium ion concentration has not been 81 86Rb+ uptake by 30% measured. Quinidine (20 pH) decreased the rate of in atrial muscle stimulated at 3.0 Hz. The decline in intracellular sodium ion concentration produced by this agent is, therefore, likely to be less than or equal to 30%. Because quinidine is expected to decrease intracellular sodium ion concentration, the effects of low sodium concentration on digitalis binding to the cardiac Na+,K+-ATPase were examined. The rate of ouabain binding was meaSured in partially-puri- 2+, ATP fied Na+,K+—ATPase from guinea pig heart in the presence of Mg and various concentrations (0-100 mM) of Na+ to compare the stimulatory effect of low and high concentrations of sodium ion on the glycoside binding. Sodium contamination due to added reagents and protein was determined to be less than 0.1 mM in the absence of added NaCl by flame photometry. The final Na+ concentrations in the reaction mixture were not adjusted for this slight error. In these experiments, choline chloride was used as an osmotic and ionic substitute for sodium. Figure 16 shows the effects of increasing Na+ concentration on the initial velocity of ATP-dependent [3H]ouabain binding to guinea pig heart Na+,K+-ATPase. From 0—5 mM NaCl, the velocity of binding de- creased as Na+ concentration was increased suggesting that low concen- trations of Na+ may inhibit rather than stimulate, glycoside binding. Similar effects of very low concentrations (:2 mM) of Na+ ion on glyco- side binding has been observed with rat brain Na+,K+-ATPase (Siegel and Josephson, 1972). Sodium concentrations in the range of intracellular Na+ ion activities measured with ion-selective microelectrodes (Cohen 31_31,, 1982), 5-15 mM, stimulated [3H]ouabain binding; however, five- fold increase in Na+ concentration from 5-25 mM produced only a 82 i: M h) l: .5 in a ATP-dependent[‘1~1]meoeiibound (p mob/inn wot/3m) 8 D {o is ioii’i‘é‘o lfld'hnMfi) (Dr 00 Figure 16. Effect of sodium ion on ATP— —dependent [ 3H]0uabain bind— ing. Partially purified guinea pig heart Na+ ,K+ -ATPase was incubated with 50 mM Tris HCl (pH 7. 5), 5 mM MgClZ, 25 mM [3 H]0uabain, 5 mM Tris -ATP (pH 6. 8) and various concentrations of NaCl. Choline chlor- ide was used as an ionic substitute for NaCl so that [Na+]+[choline+ ] 3100 mM. ATP- dependent binding was calculated as the difference in [3 H]0uabain bound during a 3 minute reaction period in the presence and the absence of ATP. Data are expressed as mean :_S. E. M. Final protein concentration is 500 ug/ml. 83 three-fold increase in [3H]ouabain binding. For comparison, rate of [3H]ouabain binding is shown in the presence of 100 mM NaCl, a concen- tration which maximally stimulates binding. From these studies, a 30% decrease in Na+ concentration, the magnitude of the maximal decrease of Na+ concentration produced by quinidine, would be expected to reduce glycoside binding by 18%. These data indicate that in 5-15 mM sodium concentrations is capable of stimulating glycoside binding to the Na+,K+-ATPase; however, the change in Na+ concentration produced by quinidine might be expected to produce only a small change in glycoside binding. ~~.——..-u____-a‘ -.. c. ‘. gen—-u. DISCUSSION A. Influence of Quinidine and Benzocaine on the Positive Inotropic Action of Digoxin 1. The Effect of Quinidine on the Inotropic Action of Digoxin In Vivo The effect of the quinidine-induced elevation in serum di- goxin concentration on the cardiac actions of digoxin in humans remains unsettled. Several papers have reported that the effects of digoxin on the heart are increased after beginning simultaneous administration of quinidine (Leahey e_t__a_1_., 1978, 1979, l980a; Belz 3331., 1982), while others report that the quinidine-induced increase in serum digoxin concentration is not accompanied by the corresponding increase in the effect of the cardiac glycoside on the heart (Hirsh 31_31,, 1980; Stein- ess 31_31,, 1980). In these studies, the glycoside effect on cardiac contractility was determined by the degree of digoxin-induced shorten- ing of the QRS complex of the electrocardiogram. Quinidine alone can prolong the QRS complex (Belz 31_31,, 1982) so that the interpretation of the effect of digoxin on contractility is complicated. Judging the effect of digoxin solely on the degree of QRS shortening without accounting for the effect of quinidine may lead to an underestimation of the magnitude of glycoside action. This may account for the lack of an increased effect of digoxin on cardiac contractility with elevated serum digoxin concentration in the studies of Steiness 31_31, (1980) 84 85 who did not measure the effect of quinidine alone, and Hirsch 31_31, (1980) who measured the effect of quinidine in only 3 of 7 individuals. In the above human studies, the effect of elevated serum digoxin concentration on the heart was estimated with indirect methods. To overcome the difficulty in measuring cardiac actions of digoxin in human subjects, studies using experimental animal models of the quini- dine/digoxin interaction have been performed. A quinidine-induced increase in plasma digoxin concentration in anesthetized guinea pigs led to a greater degree of binding of digoxin to its putative inotropic receptor in the heart, the Na+,K+-ATPase (Kim 31_31,, 1981b). In dogs, the digoxin—induced inhibition of active monovalent cation transport in cardiac ventricular tissue was increased when serum glycoside concen- tration was elevated by quinidine (Leahey 31 31,, 1980b). A recent paper from the same investigators (Warner 31_31,, 1984), however, found that the glycoside-induced inhibition of active monovalent cation transport in ventricular tissue of dogs receiving quinidine and digoxin simultaneously was not greater than the inhibition of active transport in cardiac tissue of dogs receiving the same dose of digoxin. The serum digoxin concentration in animals receiving quinidine and digoxin was, however, increased 100% compared to serum digoxin in dogs re- ceiving digoxin only. The lack of a normal dose—response relationship in dogs receiving quinidine and digoxin contrasts with the results of their earlier report (Leahey 31 31,, l980b). The reason for the dis- crepancy between the two reports is not readily apparent. The experi- mental protocols used in the two studies were different resulting in a 50% lower serum digoxin concentration (1.2 ng/ml) in dogs receiving digoxin only in the report of Warner 31_31, (1984) than in the earlier 86 study (2.5 ng/ml digoxin) of Leahey 31_31, (1980b). Simultaneous administration of quinidine increased serum digoxin concentration to 2.4 and 3.1 ng/ml in the report of Warner 31_31, (1984) and Leahey 31_ 511. (l980b), respectively. Thus, quinidine administration in the more recent report (Warner 23.21:, 1984) produced a greater increment in serum digoxin concentration so it is possible that quinidine would also have a greater effect on the pharmacologic actions of digoxin in the heart than in the earlier study (Leahey 31_31,, l980b). The conditions under which their experiments were performed appears to greatly affect their results. These studies leave open the possibility the quinidine- does interfere with the pharmacologic actions of digoxin in the heart. 2. Effects of Quinidine on the Positive Inotropic Actions of Digoxin, In Vitro Investigations of the effect of quinidine on the cardiac actions of digoxin, 1”_vivo, are complicated by changes in serum glyco- side concentration. For this reason, the present experiments were performed in isolated guinea pig left atrial muscle so that the effects of quinidine on the inotropic actions of digoxin could be observed in the absence of pharmacokinetic complications. In atrial muscle pre- parations stimulated at 1.5 Hz, quinidine in a final concentration of 3, 10 or 30 uM had no significant effects on the positive inotropic action of 0.6 uM digoxin. These results are in agreement with pre- viously published reports in isolated rat and guinea pig cardiac muscle preparations (Kim 31_31,, 1981a) and in chick heart cell cultures (Horowitz 31 31,, 1982). Williams and Mathew (1981), however, have demonstrated that prior exposure of cat papillary muscles to quinidine decreases the positive inotropic action 0f digoxin and acetylstrophan- thidin. If quinidine is added to the incubation medium at the time of or ~.:—‘-— . —- _ . 87 the maximal inotropic effect of digoxin, the positive inotropic effect of the glycoside is not decreased. Therefore, the order of drug addi- tion determined the effect of quinidine on the actions of digoxin in these experiments in their study. The reason why the order of drug addition should influence the final drug effects is unclear. Experi- ments in guinea pig atrial muscle, however, showed that quinidine addition followed by digoxin, did not decrease the inotropic actions of the glycoside. If the glycoside binding to Na+,K+-ATPase and ensuing posi- tive inotropic effect are enhanced by intracellular sodium ion, then quinidine, which reduces the rate of sodium influx, is most likely to reduce the glycoside binding and the positive inotropic effect. Thus, an apparent lack of a quinidine-induced decrease in the positive ino- tropic effect of digoxin in guinea pig atrial muscle seems to warrant further investigation. The binding of digitalis to the Na+,K+—ATPase, the proposed inotropic receptor for the cardiac glycoside, is stimulated by sodium (Matsui and Schwartz, 1968; Siegel and Josephson, 1972; Lindenmayer and Schwartz, 1973). The phosphorylated conformation of the Na+,K+- ATPase induced by sodium preferentially binds cardiac glycosides (Matsui and Schwartz, 1968; Sen 31_31,, 1969). This explains why the rate of specific ouabain binding is greater in beating heart muscle preparations stimulated at higher frequencies (Yamamoto 31_31,, 1979; Temma and Akera, 1982) and in cardiac muscle treated with batrachotoxin or monensin (Temma and Akera, 1982). Each of these experimental mani- pulations increase the rate of Sodium influx into the cardiac muscle fiber. In the absence of toxicity, the cell will maintain a relatively low intracellular sodium concentration due to the fact that the sodium 88 pump is not functioning at maximal capacity under normal conditions, and an increase in the rate of sodium influx will be matched by an increased rate of active sodium extrusion (Akera 31_31,, 1981; Eisner 31_31,, 1983b). This point is supported by numerous studies on sodium pump current (Glitsch 31_31,, l978; Eisner31 31,, 1981a) and isotopic rubidium uptake (Yamamoto 3§_31,, 1979; Akera 31_31,, 1981). An in- crease in pump activity promotes glycoside binding (Yamamoto 31_31,, 1979; Temma and Akera, 1982), presumedly by increasing the availability of the sodium-induced, glycoside-sensitive form of Na+,K+-ATPase, similar to sodium-dependent stimulation of cardiac glycoside binding to the Na+,K+-ATPase enzyme observed 13_111§3_(Matsui and Schwartz, 1968; Lindenmayer and SChwartz, 1973). Because an elevation of sodium influx or intracellular sodium ion concentration enhances the glycoside binding to Na+,K+-ATPase, the rate of glycoside binding to its receptor site is expected to be de- creased When the rate of sodium influx or intracellular sodium ion concentration is decreased. Indeed, the glycoside failed to bind to the Na+,K+-ATPase in cardiac muscle incubated in a low sodium (27 mM) buffer solution (Temma and Akera, 1983). Less dramatic decreases in extracellular sodium concentration (85 mM), delay the onset of the positive inotropic effects of cardiac glycosides (Akera 31_31,, l977). Glycoside binding under these conditions is slightly, though not signi- ficantly, slower than that seen in cardiac muscle incubated in normal sodium (145 mM) buffer solutions (Temma and Akera, 1983). Under most conditions the rate of digitalis binding to the Na+,K+-ATPase in intact cardiac muscle preparations appears to be dependent on the rate of sodium influx. 89 Quinidine is believed to exert its antiarrhythmic actions by decreasing the rate of sodium influx into cardiac cells. Quinidine has a direct depressant effect on the fast sodium current underlying the rapid upstroke of the cardiac action potential (Johnson and McKinnon, 1957; Chen 31_31,, 1975; Hondeghem and Katzung, 1980; Lee 31 31,, 1981) and on the rate of sodium uptake in beating heart muscle (Choi 33 31,, 1972). It may be hypothesized, then, that quinidine would de- crease cardiac glycoside binding to the Na+,K+-ATPase in the cardiac muscle and delay the onset of the positive inotropic effect of digi- talis under conditions where the antiarrhythmic agent depressed the rate of sodium influx. Yet, in the initial experiments, quinidine had no effect on the inotropic actions of digoxin. One possible explana- tion for the lack of the anticipated effect of quinidine is that quini- dine may not have decreased the rate of sodium influx into the cardiac cell during'membrane excitation under the conditions of the present study. To test this possibility, the effects of quinidine on the Vma 86 X of the action potential and Rb+ uptake in beating atrial muscle preparations were measured. Quinidine produced a concentration— and frequency-dependent 86 decrease in Vma and Rb+ uptake in beating atrial muscle. In pre- x parations, stimulated at 0.5 Hz, 20 uM quinidine did not have a Signi_ 86 ficant effect on Rb+ uptake and produced only a 20% decrease in Vmax’ while the same drug concentration in atrial muscle stimulated at 86 Rb+ uptake and a 50% reduction in 86 3.0 Hz produced a 30% decrease in V . The less pronounced effect of quinidine on Rb+ than Vma max may x be due to the fact that quinidine produces a prolongation of the action potential, such that a decreaSe in sodium influx during the upstroke 90 may be offset by a greater sodium influx during latter portions of the action potential. These data suggest that 20 uM quinidine had little or no effect on sodium influx rate in atrial muscle paced at 0.5 Hz but produced a significant decrease in sodium influx in preparations stimu- lated at 3.0 Hz. Quinidine is anticipated to decrease digoxin binding to the Na+,K+-ATPase in cardiac muscle when the antiarrhythmic agent decreases the rate of sodium influx. For this reason, the positive inotropic actions of digoxin and glycoside binding were measured in the presence and absence of quinidine. The positive inotropic effect of 0.6 uM digoxin was not significantly different in the presence or absence of 20 uM quinidine in atrial muscle stimulated at 0.5 Hz. The lack of an effect of quinidine under these experimental conditions was expected as the alkaloid did not decrease the rate of sodium influx. In atrial musclepreparations stimulated at 3.0 Hz, 20 uM quinidine also did not significantly influence the positive inotropic effect of 0.6 uM di- goxin, in apparent conflict with the prediction that quinidine would decrease the inotropic actions of the glycoside under conditions where quinidine decreased sodium influx rate. Fractional occupancy of the Na+,K+-ATPase by digoxin at the time of the half-maximal inotropic effect of 0.6 uM digoxin in atrial muscle stimulated at 3.0 Hz; how- ever, was significantly greater in the presence than in the absence of 20 uM quinidine. These data were unexpected because the quinidine- 86Rb+ uptake under these experimental conditions induced decrease in suggested that digoxin binding would be decreased rather than increased. The decrease in the initial velocity of [3H]ouabain binding (i.e., increase in fractional occupancy) which occurred in homogenates of 91 atrial muscle incubated with quinidine and digoxin in the absence of a significant fractional Occupancy in atria exposed to digoxin only might be interpretted to suggest that quinidine stimulated the rate of di- goxin to the Na+,K+-ATPase of the atrial muscle. This is unlikely, however, because quinidine did not significantly stimulate ouabain binding to guinea pig cardiac Na+,K+-ATPase (Figure 15) in agreement with the results of Doering (1979). Alternatively, quinidine may in- hibit [3H]ouabain binding to the Na+,K+—ATPase in homogenates. This interpretation, however, is not consistent with the present data (Figure 15) and published reports (Lowry 31 31,, 1973; Ball31_31,, 1981) which demonstrate that much higher concentrations of quinidine are required to inhibit glycoside binding to the Na+,K+-ATPase. The increase in fractional occupancy noted in atrial muscle incubated with quinidine and digoxin compared to that in atrial muscle inCubated with digoxin only, may also not be representative of true effect of 0.6 uM digoxin in the presence and absence of 20 uM quinidine. It was not possible to distinguish between these interpretations of the data or the interpre- tation that suggests the increase in fractional occupanCy represents a true increase in digoxin binding. For this reason, further binding studies were performed. Results of studies in which the previous experiment was repeated but with 2.0 uM digoxin and those in which binding was measured at the maximal inotropic effect of 0.6 uM digoxin Showed that 20 uM quinidine did not change glycoside binding to the Na+,K+—ATPase in the atrial muscle. Fractional occupancy of the glycoside-receptor sites by digoxin was also examined in atrial muscle stimulated at 3.0 Hz but incubated in a Rb+-containing buffer solution, the same buffer solution used to measure sodium pump activity. Here, 92 too, 20 pH quinidine did not significantly change the binding of 0.6 uM digoxin. Taken together, these experiments demonstrate that quini- dine did not affect the binding or the inotropic actions of digoxin, even under conditions where the alkaloid significantly decreased the rate of sodium influx into cardiac muscle. 3. Effect of Benzocaine on the Positive Inotrgpic Action of Digoxin, In Vitro The failure of quinidine to reduce digoxin binding under the condition in which quinidine inhibits sodium influx may result from an additional action of quinidine which may mask the expected effects of quinidine on the glycoside binding. In order to examine if a decrease in the rate of sodium influx affects the binding or the positive ino— tropic actions of digoxin, the effect of another drug which reduces the rate of sodium influx was studied. Local anesthetics have been shown to decrease intracellular sodium ion concentration (Deitmer and Ellis, l980a; Eisner 31 31,, l983a). Benzocaine, a local anesthetic agent, has been observed to delay the onset of the positive inotropic actions of strophanthidin in isolated canine cardiac Purkinje fibers (Bhatta- charyya and Vassalle, 1981). For this reason, the effects of benzo- caine on the inotropic actions of digoxin were examined in guinea pig left atrial muscle. The results of the action potential studies demonstrated that 300 uM benzocaine reduced V by 15% without signi- max ficantly prolonging action potential duration in atrial muscle stimu- lated at 0.5 Hz. The actions of benzocaine on \iml (Gintant 3331., x 1983) as well as INa in cardiac (Sanchez-Chapula 31_31,, 1983) and skeletal muscle cells (Schwarz 31 31,, 1977) are reported to Show little or no frequency dependence at stimulation frequencies below 7 Hz, so 93 that the action potential studies were not repeated at 3.0 Hz. The 86Rb+ uptake experiments suggested that benzocaine results of the decreased sodium influx at concentrations greater than or equal to 300 uM in atrial muscle preparations stimulated at 3.0 Hz. In contrast to the above experiments with quinidine, the onset of the positive ino- tropic effect of 0.6 uM digoxin was significantly delayed by 300 uM benzocaine in the atrial muscle. The T?72 of 0.6 uM digoxin in the absence of benzocaine was 11 minutes whereas the T??2 was 17 minutes in the presence of 300 uM benzocaine. The delay in the onset of the inotropic actions of the glycoside is similar to that observed in dog cardiac Purkinje fibers (Bhattacharyya and Vassalle, 1981). Digoxin binding in atrial muscle stimulated at 3.0 Hz, however, was not signi— ficantly slowed in the presence of 300 uM. The magnitude of the de- crease in digoxin binding which is responsible for the observed lengthen- ing of the T172 of 0.6 uM digoxin (from 11 minutes to 17 minutes by 300 uM benzocaine) is expected to correspond to approximately a 10% de- crease in the fractional occupancy of digoxin measured after incubating the atrial muscle with digoxin for 11 minutes (T??2). The expected reduction of cardiac glycoside binding produced by benzocaine is smaller than the normal variability of the data about the mean. This suggests that the local anesthetic-induced decrease in binding would be difficult to observe against normal sample variation. Alternatively, these results might indicate that benzocaine delays the onset of the positive inotropic effect of digoxin by a mechanism subsequent to glycoside binding. The concentration of benzocaine used in these experiments is similar to concentrations at which other local anesthe- tic agents have been shown to decrease calcium accumulation and release 94 by isolated sarcoplasmic reticulum (Johnson and Inesi, 1969; Nash-Alder 31_31,, 1980) and block calcium currents in Ki-depolarized cardiac muscle (Josephson and Sperelakis, 1976). It is quite possible, then, that benzocaine would delay the inotropic actions of digoxin by affect- ing the mobilization of calcium at a step subsequent to glycoside binding. Both quinidine and benzocaine have been shown to reduce the rate of sodium influx in beating cardiac muscle preparations. Although benzocaine did significantly delay the onset of the positive inotropic action of digoxin, quinidine did not significantly influence glycoside actions in the isolated atrial muscle. Furthermore, neither agent produced a significant change in glycoside binding in beating heart muscle preparations. These data do not support the hypothesis that digoxin binding and positive inotropic actions will be reduced when quinidine (or benzocaine) decreases sodium influx into the cardiac muscle fiber. Possible reasons for this discrepancy will be discussed in the next section. B. Possible Explanations for the Lack of Effect of Quinidine and Benzocaine on Digoxin Binding Quinidine does not decrease digoxin binding to atrial muscle pre— parations under conditions where the antiarrhythmic agent decreases the rate of sodium influx in the atrial muscle preparations. Similar results are observed with a local anesthetic, benzocaine. It is anti- cipated, however, that the reduction in sodium influx Would reduce the rate of glycoside binding to the Na+,K+-ATPase in cardiac muscle pre- parations. Several explanations may be possible to resolve this apparent conflict. 95 1. Effect of Quinidine and Benzocaine on Sodiun Influx into Atrial~Muscle'PreparatiOns a) Measurement of sodium influx Quinidine is proposed to decrease glycoside binding to theNa+,K+-ATPase in atrial muscle under conditions where the alkaloid decreases the rate of sodium influx. For this reason, it is important to establish conditions in which quinidine decreases the rate of sodium influx into the beating atrial muscle preparations. The maximal up- stroke velocity of the atrial muscle action potential and on-going sodium pump activity were used as estimates of the rate of sodium influx. In this project, Vma was used as an indicator of the x magnitude of the fast sodium current, though, recently, the use Of Vma x as a quantitative measure of sodium current has been questioned. For a uniform membrane action potential, Vmax should be proportional to the total ionic current crossing the membrane at the time of Vmax (Hodgkin and Huxley, 1952). The great majority of the ionic current flowing at the time of Vmax in cardiac muscle is believed to be carried by sodium ions (INa) as nonsodium currents are expected to contribute little to Vmax in the cardiac muscle action potential. As a result, Vmax has been proposed to be a good measure of INa (Hondeghem, 1978). Experi- ments measuring both Vmax and INa in the enzymatically-dispersed rat cardiac cells (Lee 31 31,, 1979; Undrovinas 31_31,, 1980) demonstrated that the observed Vmax of the action potential was similar to the calculated Vmax based upon membrane capacitance and the INa measured during voltage clamp experiments. 96 Strichartz and Cohen (1978), however, have argued that even when nonsodium currents are held at zero, Vmax is still not linear- ly related to available sodium conductance (gNa)’ where available gNa is defined by the ratio of INa to maximal INa' Even with zero nonso- dium currents, V would only be linearly related to gNa if the kine- max tics of sodium channel inactivation were much slower than channel activation so that little inactivation would have occurred at the time of Vmax‘ The model system that Strichartz and Cohen used for the cardiac action potential assumed kinetics of the fast sodium current similar to those found during voltage clamp experiments in squid axon (Cohen and Strichartz, l977). Hondeghem (1978) used a model system in which the kinetics for the sodium current were manipulated to recon- struct a ventricular muscle action potential (Beeler and Reuter, 1977), because, at the time, it was not possible to study the kinetics of the fast sodium current in the heart. The disagreements between these two model systems lies in the fact that in the squid axon sodium channel inactivation is significant at Vmax (Hodgkin and Huxley, 1952), whereas in the model of Beeler and Reuter (1977), inactivation is much slower and is minimal at the time of Vmax‘ Recent voltage-clamp experiments using enzymatically-dispersed rat heart ventricular cells (Brown 33 31,, 1981; Bodewei 31_31,, 1982) and cultured cardiac cells (Cachelin 31_31,, l983) demonstrate that kinetics of sodium channel inactivation are faster than those calculated in the model of Beeler and Reuter (1977) so that Significant inactivation may occur at Vmax‘ Experimen- tally, a nonlinear relationship between INa and Vmax has been demon- strated in rabbit Purkinje fiber (Cohen 31_31,, 1984). In this study, changes in Vmax greatly underestimated the depression of the sodium 97 current produced by tetrodotoxin or steady membrane depolarization in Purkinje fibers. The above arguments point out that the use of Vmax as an estimate of INa during the action potential upstroke remains controversial. Another source of error in relating V to INa is due max to latency changes with increasing local anesthetic drug concentration. Latency is defined as the time between the end of the electrical stimu- lus and the moment of Vmax' An increase in latency results from a stimulus which raises the membrane potential above the thresold voltage more slowly. This allows additional time for sodium channel inactiva- tion to occur before Vmax’ so that Vmax may appear smaller than if latency is held constant (Walton and Fozzard, 1979). Latency did increase in the presence of both quinidine and benzocaine (see Figures 4 and 6), especially at high drug concentrations. Thus, the drug- induced decrease in Vmax’ in part, may result from an increase in stimulus latency. Overall, the above discussion points out that the actions of quinidine or benzocaine on Vma should not be regarded as x quantitative indicators of drug effect on sodium influx rate during the action potential upstroke; the reduction in Vmax may underestimate the actual reduction in the rate of sodium influx. Nevertheless, these data should be viewed as relative measures of the effect of the drug at various concentrations and/or stimulation frequencies. Is ouabain-sensitive rubidium-uptake in beating heart muscle a good measure of sodium influx? This depends upon two factors: 1) that the sodium pump is the only mechanism for sodium extrusion in the cardiac muscle cell and 2) that 86Rb+ uptake is an accurate measure of sodium extrusion by the pump. The sodium pump is believed to be the 98 major mechanism for maintaining intracellular sodium and potassium ion concentrations (Skou, 1957; Post 31_31,, 1960). Cardiac glycosides can block active monovalent cation transport across the cell membrane (Schatzmann, 1957; Post 31_31,, 1960) and blockade of the pump by removing extracellular potassium or by toxic concentrations of cardiac glycosides causes a rapid increase in intracellular sodium ion concen— tration (Ellis, 1977; Eisner 31_31,, l98la,b). This indicates that the sodium pump is the major mechanism for maintaining low intracellular sodium ion concentration. The sodium pump, however, is not the only route for sodium efflux from the cardiac cell. The sarcolemmal sodium/ calcium and sodium/hydrogen exchange mechanisms may also extrude sodium from the cells. A mechanism for regulating intracellular pH via trans- sarcolemmal exchange of sodium and hydrogen ion has been identified in cardiac Purkinje fibers bathed in bicarbonate-free buffer solution (Deitmer and Ellis, 1980b). The importance of this exchange mechanism on sodium flux under physiological conditions is not likely to have a substantial impact on total sodium influx as the hydrogen ion concen- tration at physiological pH range is submicromolar (0.1 uM). Even assuming substantial intracellular buffer capacity, the net sodium flux for sodium/hydrogen ion exchange is likely to be very small. Increas- ing extracellular calcium concentration has been Shown to increase the rate of sodium efflux from isolated tissue preparations (Deitmer and Ellis, 1978). The magnitude of this mode of sodium/calcium exchange (calcium in and sodium out) is not well characterized in the heart. More appropriate, however, is whether the sodium flux via sodium/cal- cium exchange remained constant between the various manipulations; stimulation frequency and drug treatment. Increasing stimulation rate 99 is believed to augment calcium influx (sodium efflux) by the exchange mechanism. Faster driving rates have been shown to increase intra- cellular sodium iOn concentration (Cohen 31_31,, 1982) which could increase sodium/calcium exchange. A linear relationship between intra— cellular sodium ion activity and tension development in cardiac Pur- kinje fibers (Eisner and Lederer, 1980; Eisner 31_31,, l981a) pre- sumedly reflects increased calcium influx in exchange in intracellular sodium. This phenomenon may underlie the positive force-frequency relationship observed in many mammalian cardiac muscle preparations (Langer, 1967). Increasing stimulation frequencies, then, might be expected to enhance sodium efflux from the cardiac cell by mechanisms other than the sodium pump. Nevertheless, sodium pump activity, as measured by 86Rb+ uptake (Figure l and Akera 31_31,, 1981), is roughly proportional to stimulation rate, so that whatever the contribution of the sodium/calcium exchange mechanism to net sodium efflux, it does not appear to dramatically alter the relationship between sodium influx and sodium pump activity. The effects of quinidine-like agents on the relative importance of sodium efflux unrelated to the sodium pump is not known. Thus, the sodium pump may be considered as the major mecha- nism for sodium extrusion in cardiac muscle cells. 86 The second criterion for Rb+ uptake to be a measure of sodium influx is that isotopic rubidium uptake should accurately esti— 86Rb+ mate sodium extrusion by the sodium pump. For ouabain-sensitive uptake to estimate sodium extrusion, the coupling ratio between sodium and rubidium must remain constant. Experiments in isolated canine (Gadsby, 1980) and sheep (Eisner 31_31,, l981a,b) Purkinje fibers have demonstrated that when the sodium pump is reactivated by potassium 100 (Gadsby, 1980) or rubidium (Eisner 31_31,, l981a,b) after a Short period of sodium-loading in a potassium-free buffer, a glycoside— sensitive outward current can be measured by voltage-clamp techniques. This current decays in a monoexponential fashion with a rate constant that is dependent on the potassium (rubidium) concentration. Analysis of this "sodium pump" current has demonstrated that the coupling ratio for sodium and potassium transport by the sodium pump is fixed under a variety of cation concentrations. Other modes of sodium pump transport such as uncoupled sodium ion efflux or potassium:potassium (rubidium) exchange, however, could distort the relationship between measured 86Rb+ uptake and sodium extrusion. Uncoupled sodium efflux which is ouabain-sensitive, has been demonstrated in erythrocytes in the absence of extracellular potassium and sodium (Robinson and Flashner, 1979); however, this mode of sodium pump activity will be negligible under the present experimental conditions. Potassiumzpotassium exchange is also ouabain-sensitive (Glynn and Luthi, 1968) and it is expected to be increased at low intracellular sodium concentrations. Nevertheless, potassium:potassium exchange in erythrocytes is inhibited almost com- pletely at intracellular sodium concentrations greater than 4 mM (Simons, 1974). Because intracellular sodium concentration is higher than 4 mM, this exchange mechanism was probably insignificant in the present studies. The effect of quinidine and local anesthetics on the coupling ratio of the sodium pump has not been studied. Quinidine has 86 + 22 significant effects on active transport of Rb and Na+ by erythro- cyte membranes (Lowry 31_31,, 1973; Ball 31_31,, 1981) but only at concentrations much higher than those used in the present experiments. 86 Thus, ouabain-sensitive Rb+ uptake should be an accurate measure of the rate of sodium extrusion by the sodium pump and is a good estimate 101 of the rate of sodium influx into beating atrial muscle preparations under steady-state conditions. b) Effect of quinidine and benzocaine on sodium influx rate Quinidine produced a dose-dependent decrease in the Vmax of the guinea pig left atrial muscle which was enhanced at more rapid rates of stimulation. This data is in agreement with the fre- quency-dependent depression of Vmax observed in guinea pig ventricular muscle (Johnson and McKinnon, 1957; Chen 31_31,, 1975; Hondeghem and Katzung, 1980). And although a rate-dependent block of INa by quini- dine has not been demonstrated, lidocaine and its structural analogues which also produce frequency-dependent depression of Vmax of the cardiac action potential, cause a frequency-dependent block of INa in isolated nerve preparations (Courtney, 1975) and use-dependent block in nerve (Hille, 1977) and cardiac Purkinje fiber (Bean 31_31,, 1983) preparations. The increased effect of quinidine on Vmax at faster driving rates, then probably reflects a greater depression of INa in the atrial muscle. These results are in agreement with the "modulated receptor" hypothesis of sodium current blockade by antiarrhythmic agents (Hondeghem and Katzung, 1984). The present experiments demonstrated that quinidine reduced sodium pump activity in beating atrial muscle preparations in a dose- and frequency-dependent manner, similar to the drug's effect on action potential Vma That the quinidine produces a Similar shift in x' two independent measures of sodium influx, strengthens the view that 86 the reduction in Vmax and ouabain-sensitive Rb uptake reflected a decrease in the rate of sodium influx. 102 86 The drug-induced decrease in Rb+ uptake is not paral- leled by a reduction in Vmax of similar magnitude. Quinidine decreases 86Rb+ uptake only at concentrations which substantially decrease Vmax' For instance, in atrial muscle stimulated at 1.5 Hz, sodium pump acti- vity was reduced approximately 25% by 30 uM quinidine, a drug concen- tration which decreased Vmax 50%, yet 3 and 10 uM quinidine did not decrease 86Rb+ uptake even though Vmax was reduced 15 and 30%, respec— 86 tively. Benzocaine also produced a dose-dependent decrease in Rb+ uptake, but, unlike quinidine, the decrease in sodium pump activity was observed at a concentration (300 uM) which only minimally decreased Vmax' 86 + The difference in the relative changes in Vmax and Rb uptake may arise for the following reason: Vmax is an estimate of peak in sodium current (influx) during the action potential upstroke whereas 86Rb+ uptake is a time average of net sodium influx. ouabain-sensitive Measurements of Vmax are insensitive to changes in sodium influx occurring at times other than that during the upstroke. Conversely, 86Rb+ uptake is apt to be less sensitive to changes in sodium influx during the upstroke, particularly if sodium influx at other times during cardiac muscle excitation is significant. Quinidine has been shown to decrease the calcium current in cat ventricular papillary muscle at the concentration of lO-4M (Nawrath, 1981); however, in the concentrations used in the present experiments, quinidine may decrease Vmax without blocking other mechanisms of sodium influx such as the calcium current. Quinidine may even increase sodium influx occurring after the upstroke because it prolongs action potential duration. Benzocaine, on the other hand, is used at much higher concentrations 103 (300 uM), similar to those at which other local anesthetics have been shown to block the calcium current (Josephson and Sperelakis, 1976). This suggests that quinidine may reduce sodium influx specifically by blockade of the sodium current whereas benzocaine may reduce sodium influx in a more nonspecific manner. If this is the case, quinidine might need to produce a substantial decrease in V before a signifi- max 8°Rb+ uptake cant reduction in total sodium influx, measured by occurred, whereas the effect of benzocaine on total sodium influx might be greater than its reduction of sodium current alone. These argu— ments, however, are very speculative because, not only have mechanisms of sodium influx into cardiac atrial muscle fibers not been discretely identified, but the comparison of drug effects on action potential Vmax is reliable only as a qualitative variable in 86Rb+ are tentative as Vmax these studies. In spite of this, quinidine depresses Vmax and uptake at similar concentrations, both measures of sodium influx rate demonstrated a rate-dependent effect. These results are in agreement with those of Lee 31_31, (1981) who Showed that quinidine decreases INa in single rat cardiac myocytes. Benzocaine also reduced Vmax and 86Rb+ uptake at similar concentrations, though these concentrations are approximately ten-fOld greater than the effective concentrations of quinidine. Thus, the data indicated that these agents did decrease the rate of sodium influx in atrial muscle fibers. That substantial sodium influx occurs at times during the cardiac action potential other than during the upstroke is un— settled. In Purkinje fibers, tetrodotoxin Significantly decreases action potential duration at concentrations which have little effect on 104 Vmax (Coraboeuf 31_31,, 1979; Elharrar 31_31,, 1984). The decrease in action potential duration has been attributed to the block of a steady- state inward "window" current carried by sodium ions (Attwell 31_31,, 1979). The existence of the window current may vary between different types of cardiac muscle fibers as tetrodotoxin has little effect on action potential duration or background currents in guinea pig ventri- cular myocytes (J. Hume, personal communication). Sodium may also enter the muscle fiber via calcium channels. Reuter and Scholz (1977) calculated the ionic permeabilities the calcium channel in cow ventricular trabeculae by using the constant field equations of Goldman, Hodgkin and Katz (Hodgkin and Katz, 1949) which predict that the reversal potential of a membrane current is the product of the permeability and concentration gradient of each ion species which contributes to the current. Using these equations, the permeability of the slow current for calcium was calculated to be 100 times that for sodium, yet because extracellular sodium concentration (150 mM) was much greater than extracellular calcium concentration (1.8 mM), sodium ions might account for 25% of the total current. When extracellular sodium was decreased the expected shift in the reversal potential predicted by the constant field equation was observed. The reversal potential for the calcium current, however, may have been less positive than predicteddue to contamination of the calcium currents by outward potassium currents and the presence of a large series resis- tance in the voltage clamp circuit. In fact, recent experiments in guinea pig ventricular myocytes have demonstrated that reversal of current flow through the calcium channel occurs at +50 to +70 mV (Lee and Tsien, 1982, 1984). If this is the case, the calculated sodium 105 permeability of Reuter and Scholz (1977) may be too high. Further- more, the current carried by sodium ions through the calcium channel in a calcium-free solution has been shown to be substantially reduced by low concentrations of calcium ion (300 uM) in guinea pig ventricu- lar myocytes (Hess and Tsien, 1984). A recent paper by Matsuda and Noma (1984) also demonstrated that the removal of extracellular sodium did not have a Significant effect on peak calcium current in guinea pig ventricular myocytes (in 1.8 mM calcium buffer solution). These observations indicate that at the calcium concentration (1.2 mM) in the present experiments, sodium influx via the calcium channel may not make a substantial contribution to the calcium current. In spite of this, sodium ions entering through the calcium channel have been reported to be significant as stimulation of the sodium pump current by repetitive trains of stimuli can be reduced by D-6OO (NOTE: 0- 600 may inhibit Na current at concentrations greater than 10 uM; Dayer 31 31,, 1975), a calcium channel antagonist (Falk and Cohen, 1983), which suggests that sodium influx via the calcium channels is Signi- ficant. Recent experiments by January and Fozzard (1984) suggest that inactivation of the sodium channel, which was caused by maintaining the resting membrane potential at -50 mV in sheep Purkinje fibers, prevents the increase in intracellular sodium ion concentration which can be observed after rapid stimulus trains in Purkinje fibers with a more negative resting membrane potential. This suggests that sodium influx via non-sodium currents does not contribute to the increase in intracellular sodium ion concentration during the frequent stimula- tions. The reasons for the discrepancy between the results of Falk and Cohen (1983) and those of January and Fozzard (1984) are not l06 clear, although Purkinje fibers from different species were used and their protocols were also different. Thus, it is possible that sodium influx occurs via the slow calcium channels; however, the amount of sodium entering the cell via this pathway is unknown. If either inhibition of the slow calcium channels by D-6OO or inhibition of the sodium channels by the partial depolarization is sufficient to marked- ly reduce the stimulus induced intracellular sodium ion accumulation, one may conclude that both mechanisms of the sodium influx are substan- tial and contribute to the accumulation of sodium ions observed during a rapid stimulus train. The present experiments also suggest that sodium ions enter the cardiac cell at times other than the upstroke or plateau phase of the action potential. Sodium pump activity is beat-depend- 86 ent, yet in quiescent atria, ouabain-sensitive Rb+ uptake is 3.9 nmoles/mg wet weight/20 min (Figure l), which is about 35% of the 86Rb+ uptake measured at 1.5 Hz. A substantial ouabain-sensitive basal sodium leak is also suggested by measurements of the atrial muscle resting potential. The average resting potential, -78 mV, is less than the calculated potassium reversal potential, EK, of -86 mV (assuming intracellular potassium concentration equals 150 mM) for the present experiments. This leviation of the resting potential from EK is believed to be due to a low permeability of the resting membrane to sodium ions (Sperelakis, 1979). Thus, even in the absence of electri- cal stimulation, a baseline sodium leak still occurs. Electrical stimulation might, then, increase sodium influx via the sodium channel and other ionic channels. Studies in beating rabbit ventricular muscle have also shown that the rate of sodium efflux measured by labelled sodium ion exchange studies may exceed the amount of sodium lO7 influx which occurs during the fast sodium current (Langer, 1974). Thus, net influx in the heart muscle may be the sum of sodium entering via the sodium channel and other pathways. This may account for dis- crepancy between the effects of quinidine and benzocaine on Vmax and 86Rb+, as the two drugs may have different effects on sodium influx which occurs at other times than the upstroke of the action potential. In spite of the difference in the effects of quinidine 86 and benzocaine on Vma and Rb+ uptake, both drugs appear to decrease x the rate of sodium influx in atrial muscle preparations under specific experimental conditions. Digoxin binding, however, is not significant- ly influenced by either drug under these specific experimental condi- tions. Thus, the lack of quinidine or benzocaine-induced effect on glycoside binding cannot be attributed to the absence of a decrease in sodium influx produced by quinidine or benzocaine. 2. Effect of the Positive Inotropic Action of Quinidine on Digoxin Binding_ Another possible explanation for the lack of a quinidine- induced decrease in digoxin inotropic effect is that the positive inotropic effect of quinidine itself, may be responsible for changing glycoside binding so that a decrease in sodium influx would not produce a decline in total digoxin binding to the Na+,K+-ATPase. In guinea pig left atrial muscle, quinidine produced a positive inotropic effect. At higher concentrations, a biphasic effect on developed tension was observed. The effects of quinidine on contractility of cardiac muscle appear to be species and tissue—dependent. Quinidine produces a nega- tive inotropic effect in dog (Folle and Aviado, l966; Pruett and Woods, 1967) and guinea pig ventricular muscle (Kim et_al,, l98la); however, it produces a positive inotropic effect in rat atrial and ventricular l08 muscle (Kim gt al,, l98la) and ferret papillary muscle (Lash gt_al,, l982). Nawrath (1981) also reported that quinidine produces a positive inotropic effect in guinea pig left atrial muscle which is not in- fluenced by atropine. The positive inotropic effect was then followed by a decline in developed tension. The positive inotropic effect of quinidine observed in the present study, therefore, is in agreement with earlier published reports. It is unlikely that the change in baseline contractility of the atrial muscle produced by quinidine altered the binding of digoxin. Changing extracellular calcium concentration from l.25 to 2.5 mM, which increases force development more than two-fold, does not change the rate of onset of the positive inotropic action of ouabain in guinea pig left atrial muscle (Akera et_al:, l977). Furthermore, although in- creasing stimulation frequency from 0.5 to 2.0 Hz increases the rate of glycoside binding to the Na+,K+-ATPase in beating atrial muscle, the extent of binding at peak inotropic effect is not significantly differ— ent between those preparations driven at 0.5 Hz and those at 2.0 Hz (Temma and Akera, l982). These data suggest that merely changing the baseline inotropic state of the cardiac muscle will not alter glycoside action unless sodium influx is also changed. The positive inotropic action of quinidine does not appear to be due to an increase in sodium influx, as measured by 86 Rb+ uptake. Because quinidine produces a positive inotropic effect in some cardiac muscle preparations and a negative inotropic effect in others, it is possible to examine the influence of the inotropic actions of quinidine on the digoxin-induced increase in cardiac contractility. In cardiac muscle preparations in which quinidine has a positive inotropic effect (Kim §t_al,, 198la) l09 and a negative inotropic effect (Kim §t_al,, l98la; Horowitz et al,, l982), the positive inotropic effect of digoxin and the effect of glycoside-induced inhibition of the sodium pump has been found to be unchanged in the presence of quinidine. Furthermore, in guinea pig ventricular muscle, 6 uM quinidine does not influence the positive ino- tropic effect or the extent of binding of digoxin under experimental conditions very similar to those employed in the present study. Quini- dine (6 uM) also produced a 35 percent decrease in force development in these preparations (Kim §t_§l,, l98la). It is unlikely that the binding of digoxin or the positive inotropic effect of the glycoside would be altered due to the inotropic actions of quindine. 3. Effect of Quinidine on the Fractional Occupancy of Digoxin in Rubidium-Containing Buffer Solutions Those experiments which examined the effect of quinidine on the inotropic actions and binding of digoxin were performed in a bicar- bonate buffer containing 5.8 mM K+, while sodium pump studies were carried out in a buffer solution containing 5 mM Rb+. Rubidium acts as an ionic substitute for K+ but it has greater affinity for the Na+,K+- ATPase than K+ (Post et_al,, l972) and membrane permeability for Rb+ is less than that for K+ (Sjodin, l959; Hagiwara gt al,, l972). For this reason, the effect of quinidine or benzocaine on 86Rb+ uptake might not accurately reflect the actions of these two agents on sodium pump acti- vity in K+-containing solutions. Fractional occupancy of the Na+,K+- ATPase by digoxin in atrial muscle preparations stimulated at 3 Hz and incubated in a 5 mM Rb+-containing buffer solution, however, was not significantly influenced by 20 uM quinidine, similar to the lack of effect of quinidine on digoxin binding in atrial muscle incubated in llO K+-containing buffer solutions. These results confirm the finding that quinidine does not influence digoxin binding under conditions where quinidine decreases the rate of sodium influx. Thus, the failure of the present finding to support the hypothesis which suggests glycoside binding will be decreased when quinidine reduces sodium influx cannot be resolved as difference in effects of quinidine on the rate of sodium influx under the different experimental protocols used to measure glyco- side binding and the rate of sodium influx. 4. Quinidine-induced Stimulation of Digoxin Binding to Na+,K+- ATPase Quinidine fails to significantly alter digoxin binding under conditions in which the quinidine appears to decrease sodium influx rate in atrial muscle preparations. The lack of an effect on digoxin binding raises the possibility that the decrease in digoxin binding which does occur due to a reduction in sodium influx rate is offset by a second effect of quinidine to stimulate glycoside binding. The net effect then will be that cardiac glycoside binding is not changed in the presence of quinidine. A stimulation of digoxin binding can occur by increasing the apparent affinity of digoxin for the Na+,K+-ATPase or by stabilizing the glycoside-sensitive form of the enzyme. An in- crease in the apparent affinity of the Na+,K+—ATPase for digoxin can result either from a decrease in the rate of dissociation or an increase in the rate of association of the glycoside. The dissociation of bound digoxin from lamb kidney Na+,K+-ATPase is reported to be insensitive to even millimolar concentrations of quinidine (Ball et_al,, l98l). At quinidine concentrations greater than or equal to 300 uM, the rate of ouabain binding was decreased in the presence or absence of 0.6 mM lll K+, consistent with published reports on the effect of the alkaloid on cardiac glycoside binding to the Na+,K+-ATPase (Lowry gt_al,, 1973; Ball et_al,, l98l). A slight stimulation (5-10 percent) of the rate of [3H]ouabain binding was observed with 3 and l0 uM quinidine when the binding reaction was performed in the presence of 0.6 mM K+. This stimu- lation, though not statistically significant, may indicate that a simi- lar 5-lO percent stimulation of digoxin binding could occur in beating atrial muscle preparations. The magnitude of this stimulation, however, does not appear to be large enough to offset the expected quinidine- induced inhibition of glycoside binding. These results indicate that a quinidine-induced stimulation of glycoside binding to the Na+,K+- ATPase would not account for a lack of the anticipated quinidine/digoxin interaction in guinea pig atrial muscle preparations. 5. Sodium Dependence of Glycoside Binding to the Na+,K+-ATPase The absence of a quinidine-induced stimulation of ouabain binding to the Na+,K+-ATPase suggests that the relationship between the availability of glycoside-sensitive form of the sodium pump and glyco- side binding will also remain unchanged in the presence of quinidine. The working hypothesis suggests that the availability of the glycoside— sensitive form of the sodium pump is proportional to the rate of sodium influx under steady state conditions. Quinidine is, therefore, ex- pected to decrease digoxin binding to the sodium pump in atrial muscle fibers under conditions where the antiarrhythmic agent decreases sodium influx: yet, the results do not fulfill this expectation. For this reason, the working hypothesis was re-examined. ll2 The concept that cardiac glycoside binding to the Na+,K+- ATPase in the cardiac cell is determined by the rate of sodium influx. is formulated from the data showing that digitalis binding and positive inotropic actions in cardiac muscle tissue are enhanced by maneuvers which increase the rate of sodium influx, such as rapid stimulation (Park and Vincenzi, l975; Bentfeld et_al,, 1977; Yamamoto §t_al,, 1979; Temma and Akera, l983) and the treatment with agents which increase passive sodium leak across the sarcolemma (Akera §t_al,, l977; Temma and Akera, l982). Conversely, incubation of cardiac muscle preparations in buffer solutions with moderate reductions in sodium concentration slows the onset of the positive inotropic effect of ouabain (Akera gt. al,, 1977) and a further reduction to less than 30 mM sodium, abolishes the positive inotropic effect (Linden and Brooker, T980; Wiggins and Bentolila, l980; Temma and Akera, T983) and specific binding (Temma and Akera, l983) of cardiac glycosides. The above concept is also con- sistent with the fact that sodium enhances glycoside binding to the 2+ and ATP (Matsui and isolated Na+,K+-ATPase in the presence of Mg Schwartz, l968; Lindenmayer and Schwartz, l973), ligand conditions similar to those believed to promote binding in the beating heart (Akera gt_al,, l976b). These binding studies, though, used sodium con- centrations considerably higher than those measured in quiescent and beating heart muscle preparations (4-15 mM) with sodium-sensitive microelectrodes (Ellis, l977; Cohen §t_al,, l982; January and Fozzard, l984). The results of the present experiments which assayed ATP- dependent ouabain binding to Na+,K+-ATPase isolated from guinea pig heart showed that at sodium concentrations greater than 5 mM, ouabain ll3 binding was stimulated with increasing sodium concentration. The con- centration of sodium to half-maximally stimulate ouabain binding in the absence of K+ has been reported to be l3-20 mM (Siegel and Josephson, l972; Lindenmayer and Schwartz, l973). Sodium stimulation of ouabain binding to Na+,K+-ATPase obtained from guinea pig heart does occur at concentrations similar to those measured in cardiac muscle cells. Furthermore, quinidine does not appear to alter the relation between sodium concentration and glycoside binding, as [3H]ouabain binding to guinea pig microsomal Na+,K+-ATPase in the presence of 20 mM sodium, a concentration which will half-maximally stimulate glycoside binding, is not influenced by quinidine except at concentrations greater than 100 “M (Figure l5). A decrease in intracellular sodium ion concentration produced by quinidine would then be expected to decrease the rate of glycoside binding to the Na+,K+-ATPase as predicted by the working hypothesis. The fact that quinidine and benzocaine did not change glyco- side binding suggests several possible explanations to reconcile the experimental data and the working hypothesis. First, the hypothesis that a quinidine-induced decrease in the rate of sodium influx will decrease the rate of cardiac glycoside binding to the Na+,K+-ATPase may be unsubstantiated. As pointed out previously, however, a large body of experimental evidence supports the hypothesis that the rate of car- diac glycoside binding to the Na+,K+-ATPase in intact cardiac muscle is dependent on the rate of sodium influx into the heart cell. Alterna- tively, the lack of an effect of quinidine on digoxin binding even under conditions where the alkaloid decreased the rate of sodium influx may suggest that the stimulation of glycoside binding by sodium cannot ll4 be explained only by postulating that the rate of sodium influx determ- ines the rate of digitalis binding. A reduction of extracellular sodium concentration to 85 mM decreases the rate of onset of the posi- tive inotropic action of digitalis (Aker et_al,, 1977) and slightly reduces glycoside binding (Temma and Akera, 1983), while lowering extracellular sodium concentration to less than 30 mM abolishes the positive inotropic effects of cardiac glycosides (Linden and Brooker, l980; Wiggins and Bertolila, l980; Temma and Akera, l983). Yet, quini- dine and benzocaine do not influence digoxin binding under conditions where they decrease the rate of sodium influx by 25-30%. Reduction of extracellular sodium concentration to 50 mM from 145 mM also reduces 86Rb+ uptake by 30% in guinea pig left atrial muscle stimulated at l.5 Hz (Yamamoto et_al,, 1979). These data point out that although the reduction in the sodium pump activity is similar, the effect of de- creased extracellular sodium on glycoside binding is greater than the effect of either quinidine or benzocaine. This implies that the con- centration of extracellular, as well as intracellular, sodium may be important in determining the rate cardiac glycoside binding to the sodium pump. Similar schemes have been proposed to explain high and low affinity binding sites for sodium ion stimulation of digitalis binding in squid axon (Baker and Willis, 1972) and isolated Na+,K+- ATPase (Inagaki gt al,, 1974). In brief, extracellular sodium ion may stabilize the conformation of the Na+,K+-ATPase, which binds the car— diac glycoside, or stabilize the glycoside-enzyme complex. Thus, when extracellular sodium ion concentration is reduced, the reduction in the stabilizing effects due to reduced extracellular sodium ion concentra- tion and the reduced stimulatory effect of intracellular sodium ions may ll5 act additively. If this were the case, the effect of quinidine or benzocaine, which reduces intracellular sodium only, is anticipated to be smaller. Finally, the difference in the fractional occupancy of the Na+,K+-ATPase by digoxin in the presence and absence of quinidine (or benzocaine) may be too small to be distinguished from sample varia- tion. Quinidine and benzocaine decrease the rate of sodium influx by 25-30%, so that the maximal decrease in intracellular sodium ion con- centration would be 30%, assuming the rate of sodium extrusion by the sodium pump remains unchanged. However, sodium pump activity decreases as intracellular sodium ion concentration falls (Eisner §t_glf, l981a,b), so that the actual decrease in intracellular sodium ion concentration is expected to be less than 30%. A concentration of procaine (0.l mM), which decreases the rate of sodium influx by 30%, reduces intracellular sodium ion concentration less than l0% in quiescent Purkinje fibers (Deitmer and Ellis, l980a). The sodium-induced stimulation of [3H]- ouabain binding to guinea pig cardiac Na+,K+-ATPase suggests that a lO—30% reduction in sodium would reduce glycoside binding by 6-l8%. Digoxin binding to the Na+,K+-ATPase has a similar sodium dependence as ouabain binding (Fricke and Klaus, l978), which suggests that a simi- lar reduction in digoxin binding would occur with a 30% reduction in sodium influx rate in atrial muscle preparations. Sample variation was typically lO-15% around the mean in fractional occupancy experiments. Thus, a change in fractional occupancy much larger than that which is expected would be required to discern a significant effect on binding. Therefore, it is possible the decrease in binding produced by quinidine and benzocaine is too small to be seen against sample variation. ll6 Furthermore, this small decrease in binding may not be pharmacologi- cally significant. This may explain why quinidine does not influence the positive inotropic actions of digoxin in the guinea pig left atrial muscle preparations. In conclusion, the influence of quinidine on the positive ino- tropic effects of digoxin was examined in isolated left atrial muscle of the guinea pig heart under a variety of experimental conditions. Under none of the experimental conditions examined was the expected interaction between quinidine and digoxin observed. The rate of digoxin binding to the Na+,K+-ATPase in the atrial muscle and the resulting positive inotropic effect were unchanged in the presence of quinidine, even in experimental conditions where the quinidine decreased sodium influx rate. These results suggest that in the clinical setting, the rise in serum digoxin concentration in the presence of quinidine will be accompanied by a greater pharmacological effect of the glycoside. These findings indicate that the "target" concentration for digoxin in plasma should not be altered when the combination of digoxin and quinidine is to be administered. BIBLIOGRAPHY BIBLIOGRAPHY Akera, T.: Effects of cardiac glycosides on Na+,KI-ATPase. In Hand- book of Experimental Pharmacology, Part I, Vol. 56, ed. by K. Greeff, l98l, Springer-Verlag, Inc., Berlin, pp. 287-336. Akera, T. ,Baskin, S. 1., Tobin, T. and Brody, T. M. Ouabain: Tempo- ral relationship between the inotropic effect and the in vitro binding to, and dissociation from, Na+ ,K+ -activated ATPase. Naunyn- Schmiedeberg' 5 Arch. Pharmacol. gzz_: 151-162, 1973. Akera, T. Bennett, R. T., Olgaard, M. K. and Brody, T. M. Cardiac Na+ ,KT-adenosine triphosphatase inhibition by ouabain and myo- cardial sodium: A computer simulation. J. Pharmacol. Exp. Ther. 1993 287-297, l976a Akera, T. and Brody, T.M.: Membrane adenosine triphosphatase: The effect of potassium on the formation and dissociation of the ouabain-enzyme complex° J. Pharmacol. Exp. Ther. lZ§5 545-557, l97l. Akera, T., Ku, D., Tobin, T. and Brody, T.M.: The complexes of ouabain with sodium- and potassium-activated adenosine triphosphatase formed with various ligands: Relationship to the complex formed in the beating heart. Mol. Pharmacol. lg; lOl-ll4, l976b Akera, T., Larsen, F.S. and Brody, T.M.: Correlation of cardiac sodium- and potassium-activated adenosine triphosphatase activity with ouabain-induced stimulation. J. Pharmacol. Exp. Ther. 1133 145-151, 19700 Akera, T., Olgaard, M.K., Temma, K. and Brody, T.M.: Development of the positive inotropic action of ouabain: Effects of transmem- brane sodium movement. J. Pharmacol. Exp. Ther. ggg; 675-684, 1977. Akera, T., Yamamoto, 5., Temma, K., Kim, D-H. and Brody, T.M.: Is ouabain-sensitive rubidium or potassium uptake a measure of sodium pump activity in isolated cardiac muscle? Biochim. Biophys. Acta 559; 779-790, 1981. Albuquerque, E.X., Seyama, I. and Narahashi, T.: Characterization of batrachatoxin-induced depolarization of the squid giant axons. J. Pharmacol. Exp. Ther. 1855 308-3l4, T973. 117 118 Allen, J. C., Entman, M. L. and Schwartz, A.: The nature of the trans- port adenosine triphosphatase- digitalis complex. VIII. The relationship between in vivo- -formed (3 H- ouabain- Na+, K+ -adenosine triphosphatase) complex and ouabain- induced positive inotropism. J. Pharmacol. Exp. Ther. 1925 105-112, 1975. Allen, J.C., Martinez- Maldonado, 11. ,Eknoyan, G. ,Suki, W. N. and Schwartz, A.: Relation between digitalis binding in vivo and inhibition of sodium, potassium- -adenosine triphosphatase in canine kidney. Biochem. Pharmacol. 295 73-80,1971. Attwell, 0., Cohen, 1., Eisner, D., Ohda, M. and Ojeda, C.: The steady-state TTXsensitive (“window") sodium current in cardiac Purkinje fibers. Pfluegers Arch. 3295 137-142, 1979. Baker, P.F. and Willis, J.S.: Inhibition of the sodium pump in squid giant axons by cardiac glycosides: The dependence on extracellu- lar ions and metabolism. J. Physiol. (Lond.) 2245 463-475, 1972. Ball, W.J., Jr., Tse-Eng, D., Wallick, E.T., Bilezikian, J.P., Schwartz, A. and Butler, V.P., Jr.: Effect of quinidine on the digoxin receptor in vitro. J. Clin. Invest. 685 1065-1074, 1981. Bean, B.P., Cohen, C.J. and Tsien, R.W.: Lidocaine block of cardiac sodium channels. J. Gen. Physiol. 815 613-642, 1983. Beeler, G.W. and Reuter, H.: Reconstruction of the action potential of ventricular myocardial fibres. J. Physiol. 2685 177-210, 1977. Belz, G.G., Doering, W., Aust, P.E., Heinz, M., Matthews, J. and Schneider, 8.: Quinidine-digoxin interaction: Cardiac efficacy of elevated digoxin concentration. Clin. Pharmacol. Ther. 315 548-554, 1982. Bentfeld, M., Lullman, H., Peters, T. and Proppe, D.: Interdependence of ion transport and the action of ouabain in heart muscle. Brit. J. Pharmacol. 615 19-27, 1977. Besch, H.R. Jr., Allen, J.C., Glick, G. and Schwartz, A.: Correlation between the inotropic action of ouabain and its effects on sub- cellular enzyme systems from canine myocardium. J. Pharmacol. Exp. Ther. 1115 112, 1970. Besch, H. R. Jr. and Watanabe, A. M. Binding and effect of tritiated quinidine on cardiac subcellular enzyme +systems: Sarcoplasmic reticulum vesicles, mitochondria and Na+ ,K+ -adenosine triphos- phatase. J. Pharmacol. Exp. Ther. 292_: 354— 364, 1977. Besch, H.R. Jr. and Watanabe, A.M.: The positive inotropic effect of digitoxin: Independence from sodium accumulation. J. Pharmacol. Exp. Ther. 2925 958-965, 1978. 119 Bhattacharyya, M.L. and Vassalle, M.: The effect of local anaesthe- tics on strophanthidin toxicity in canine cardiac Purkinje fibres. J. Physiol. 3125 125-142, 1981. Biedert, 3., Barry, W.H. and Smith, T.W.: Inotropic effects and changes in sodium and calcium contents associated with inhibition of monovalent cation active transport by ouabain in cultured myocardial cells. J. Gen. Physiol. 745 479-494, 1979. Bigger, J.T. Jr.: The quinidine-digoxin interaction. New Engl. J. Med. 3915 779-781, 1979. Bigger, J.T. Jr. and Hoffman, B.F.: Antiarrhythmic drugs. In Ih§_ Pharmacological Basis for Therapeutics, A. Goodman, L.S. Gilman and A. Gilman (eds.), 1980, MacMillan Publishing Co., New York, pp. 761-792. Bodewei, R., Hering, S., Lemke, B., Rosenshtraukh, L.V., Undrovinas, A.I. and Wollenberger, A.: Characterization of the fast sodium current in isolated rat myocardial cells: Simulation of the clamped membrane potential. J. Physiol. 3255 301-315, 1982. Brown, A.M., Lee, K.S. and Powell, T.: Sodium currents in single rat heart muscle cells. J. Physiol. 3185 479-500, 1981. Cachelin, A.B., DePeyer, J.E., Kokubun, S. and Reuter, H.: Sodium channels in cultured cardiac cells. J. Physiol. 3495 389-401, 1983. - Chen, C-M., Gettes, L.S. and Katzung, B.G.: Effect of lidocaine and quinidine on steady-state characteristics and recovery kinetics of (dV/dt)max in guinea pig ventricular myocardium. Circ. Res. 375 20-29, 1975. Chen, T.S. and Friedman, H.S.: Alteration of digoxin pharmacokinetics by a single dose of quinidine. J. Am. Med. Assoc. 2445 669-672, 1980. Choi, S.J., Roberts, J. and Kelliher, G.J.: The effect of propranolol and quinidine on 22Na- and 42K-exchange in the cat papillary muscle. Eur. J. Pharmacol. 205 10-21, 1972. Clausen, T. and Hansen, 0.: Active Na-K transport and the rate of ouabain binding. The effect of insulin and other stimuli or skele— tal muscle and adipocytes. J. Physiol. 2195 415-430, 1977. Cody, R.J. Jr., Walker, A.M., Greenblatt, D.J. and Jick, H.: In- creased digoxin toxicity during quinidine administration. Circu- lation 62(Supp1. III): 183, 1980. Cohen, C.J., Bean, B.P. and Tsien, R.W.: Maximal upstroke velocity as an index of available sodium conductance. Circ. Res. 545 636-651, 1984. 120 Cohen, C.J., Fozzard, H.A. and Sheu, S-S.: Increase in intracellular sodium ion activity during stimulation in mammalian cardiac muscle. Circ. Res. 595 651-662, 1982. Cohen, I.S., Jick, H. and Cohen, S.I.: Adverse reactions to quinidin in hospitalized patients: Findings based on data from the Boston Collaborative Drug Surveillance program. Prog. Cardiovasc. Dis. 205 151-163, 1977. Cohen, 1.5. and Strichartz, G.R.: On the voltage—dependent action of tetrodotoxin. Biophys. J._lZ: 275-9, 1977. Colatsky, T.J.: Mechanisms of action of lidocaine and quinidine on action potential duration in rabbit cardiac Purkinje fibers. Circ. Res. 595 17—27, 1982. Coraboeuf, E., Debroubaix, E. and Coulombe, A.: Effect of tetrodotoxin on action potentials of the conducting system of the dog heart. Am. J. Physiol. 2365 H561-H567, 1979. Courtney, K.R.: Mechanism of frequency-dependent inhibition of sodium currents in frog myelinated nerve by lidocaine derivative GEA 968. J. Pharmacol. Exp. Ther. 1955 225-236, 1975. Dayer, R., Kaufmann, R. and Mannhold, R.: Inotropic and electrophysio- logical actions of verapamil and D600 in mammalian myocardium. Naunyn-Schmiedeberg's Arch. Pharmacol. 2995 69-80, 1975. Deitmer, J.W. and Ellis, D.: Changes in the intracellular sodium activity of sheep heart Purkinje fibres produced by calcium and other divalent cations. J. Physiol. 2125 437-453, 1978. Deitmer, J.W. and Ellis, 0.: The intracellular sodium activity of sheep heart Purkinje fibres. Effects of local anesthetics and tetrodotoxin. J. Physiol. 3995 269-282, 1980a. Deitmer, J.W. and Ellis, D.: Interactions between the regulation of the intracellular pH and sodium activity of sheep cardiac Purkinje fibres. J. Physiol. 3945 471-488, 1980b. Doering, W.: Quinidine-digoxin interaction. New Engl° J. Med. 3915 400-404, 1979. Doering, W., Hohne, M. and Wagner, J.: Chinidin-digoxin-interaction. Deutsche Med. Wochenschrift 1965 1373-1376, 1981. Doering, W. and Konig, E.: Anstieg der digoxinkonzentration im serum unter chinidinmedikation. Med. Klin. 735 1085-1088, 1978. Doering, W. and Konig, E.: Chinidin erhoht den digoxin-spiegel. Med. Klin. 765 395-398, 1981. 121 Doherty, J.E., Straub, K.D., Murphy, M.L., de Soyza, N., Bissett, J.K. and Kane, J.J.: Digoxin-quinidine interaction. Am. J. Cardiol. 455 1196-1200, 1980. Eisner, D.A. and Lederer, W.J.: Characterization of the electrogenic sodium pump in cardiac Purkinje fibres. J. Physiol. 3935 441—474, 1980. Eisner, D.A., Lederer, W.J. and Sheu, S-S.: The role of intracellular sodium activity in the anti-arrhythmic action of local anaesthe- tics in sheep Purkinje fibres. J. Physiol. 3495 239-257, 1983a. Eisner, D.A., Lederer, W.J. and Vaughan-Jones, R.D.: The dependence of sodium pumping and tension on intracellular sodium activity in voltage-clamped sheep Purkinje fibres. J. Physiol. 3175 163-187, 1981a. Eisner, D.A., Lederer, W.J. and Vaughan-Jones, R.D.: The effects of rubidium ions and membrane potential on the intracellular sodium activity of sheep Purkinje fibres. J. Physiol. 3115 189-205, 1981b. Eisner, D.A., Vaughan-Jones and Lederer, W.J.: Letter to the editor. Circ. Res. 535 834-835, 1983b. Ejvinsson, G.: Effect of quinidine on plasma concentrations of di- goxin. Brit. Med. J. 15 279-280, 1978. Elharrar, V., Atarashi, H. and Surawicz, B.: Cycle-length dependent action potential duration in canine cardiac Purkinje fibers. Am. J. Physiol. 2475 H936-H945, 1984. Ellis, D.: The effects of external cations and ouabain on the intra- cellular sodium activity of sheep heart Purkinje fibers. J. Physiol. 2735 211-240, 1977. Falk, R. and Cohen, 1.: Post-drive membrane currents in canine Purkinje fibers. Biophys. J. 335 36a, 1981. Falk, R. and Cohen, I.S.: Ionic and pharmacological dependence of post-drive current in canine Purkinje fibers. Biophys. J. 415 73a, 1983. Fichtl, B. and Doering, W.: The quinidine-digoxin interaction in perspective. Clin. Pharmacokin. 85 137-154, 1983. Fichtl, B., Doering, W. and Seidel, H.: The quinidine-digoxin inter- action in patients with impaired renal function. Int. J. Clin. Pharmaco1. 215 229-233, 1983. Folle, L.E. and Aviado, D.M.: The cardiopulmonary effects of quinidine and procainamide. J. Pharmacol. Exp. Ther. 1545 92-102, 1966. 122 Fosset, M., DeBarry, J., Lenoir, M-C and Lazdunski, M.: Analysis of molecular aspects of Na and Ca uptakes by embryonic cardiac cells in culture. J. Biol. Chem. 2525 6112-6117, 1977. Fricke, U. and Klaus, W.: Sodium-dependent cardiac glycoside binding: Experimental evidence and hypothesis. Br. J. Pharmacol. 625 255- 257, 1978. Fuchs, F., Gertz, E.W. and Briggs, F.N.: The effect of quinidine on calcium accumulation by isolated sarcoplasmic reticulum of skele- tal and cardiac muscle. J. Gen. Physiol. 525 955-968, 1968. Gadsby, D.C.: +Activation of electrogenic Na+/K+ exchange by extra- cellular K in canine cardiac Purkinje fibers. Proc. Natl. Acad. Sci. 215 4035-4039, 1980. Gardner, J.D. and Kiino, D.R.: Ouabain binding and cation transport in human erythrocytes. J. Clin. Invest. 525 1845-1851, 1973. Gibson, T.P. and Nelson, H.A.: Digoxin alters quinidine and quinidine alters digoxin pharmacokinetics. J. Lab. Clin. Med. 955 417-428, 1979. Gintant, G.A., Hoffman, B.F. and Naylor, R.E.: The influence of molecular form of local anesthetic-type antiarrhythmic agents on reduction of the maximum upstroke velocity of canine cardiac Purkinje fibers. Circ. Res. 525 735-746, 1983. Glitsch, H.G., Grabowski, W. and Thielen, J.: Activation of the electrogenic sodium pump in guinea-pig atria by external potassium ions. J. Physiol. 2765 515-524, 1978. Glitsch, H.G., Reuter, H. and Scholz, H.: The effect of the internal sodium concentration on calcium fluxes in isolated guinea-pig auricles. J. Physiol. 2995 25-43, 1970. Glynn, I.M.: The action of cardiac glycosides on ion movements. Pharmacol. Rev. 155 381-407, 1964. Glynn, I.M. and LUthi, U.: The relation between ouabain—sensitive po- tassium efflux and the hypothetical dephosphorylation step in the "transport ATPase" system. J. Gen. Physiol. 515 3855-3915, 1968. Gold, H., Modell, W. and Price, L.: Combined actions of quinidine and digitalis on the heart. Arch. Int. Med. 595 766-796, 1932. Goldman, R.H., Coltart, D.J., Friedman, J.P., Nola, G.T., Berke, D.K., Schweizer, E. and Harrison, D.C.: The inotropic effects of digoxin in hyperkalemia. Circulation 485 830-838, 1973. Hager, W.D., Fenster, P., Mayersohn, M., Perrier, D., Graves, P., Marcus, F.I. and Goldman, S.: Digoxin-quinidine interaction. New Engl. J. Med. 3995 1238-1241, 1979. 123 Hagiwara, S., Eaton, D.C., Stuart, A.E. and Rosenthal, N.P.: Cation selectivity of the resting membrane potential of squid axon. J. Membr. Biol. 95 373-384, 1972. Hess, H.H. and Pope, A.: Effect of metal cations on adenosine tri- phosphatase activity of rat brain. Fed. Proc. 155 196, 1957. Hess, P. and Tsien, R.W.: Mechanism of ion permeation through calcium channels. Nature 3095 453-456, 1984. Hille, B.: Local anesthetics: Hydrophilic and hydrophobic pathways for the drug-receptor reaction. J. Gen. Physiol. 695 497-515, 1977. Hirsh, P.D., Weiner, H.J. and North, R.L.: Further insights into digoxin—quinidine interaction: Lack of correlation between serum digoxin concentration and inotropic state of the heart. Am. J. Cardiol. 465 863-868, 1980. Hodgkin, A.L. and Huxley, A.F.: A quantitative description of mem- brane current and its application to conduction and excitation in nerve. J. Physiol. 1175 500-544, 1952. Hodgkin, A.L. and Katz, B.: The effect of sodium ions on the elec- trical activity of the giant squid axon. J. Physiol. 1985 37- 77, 1949. Hoffmann, B.F. and Bigger, J.T. Jr.: Digitalis and allied cardiac glycosides. In The Pharmacological Basis of Therapeutics, A. Goodman, L.S. Goodman and A. Gilman (eds.), 1980, MacMillan Publishing Co., New York, pp. 729-760. Hoffman, J.F.: The interaction between tritiated ouabain and the Na-K pump in red blood cells. J. Gen. Physiol. 545 3435—3505, 1969. Hondeghem, L.M.: Validity of Vma as a measure of the sodium current in cardiac and nervous tissues. Biophys. J. 235 147-152, 1978. Hondeghem, L. and Katzung, B.G.: Test of a model of antiarrhythmic drug action. Effects of quinidine and lidocaine on myocardial conduction. Circulation 615 1217-1226, 1980. Hondeghem, L.M. and Katzung, B.G.: Antiarrhythmic agents: The modu- lated receptor mechanism of action of sodium and calcium channel- blocking drugs. Ann. Rev. Pharmacol. Toxicol. 245 387-423, 1984. Horowitz, J.D., Barry, W.H. and Smith, T.W.: Lack of interaction between digoxin and quinidine in cultured heart cells. J. Phar— macol. Exp. Ther. 2295 488-493, 1982. 124 Hougen, T.J. and Smith, T.W.: Inhibition of myocardial monovalent cation active transport by subtoxic doses of ouabain in the dog. Circ. Res. 425 856-863, 1978. Hougen, T.J., Spicer, N. and Smith, T.W.: Stimulation of monovalent cation active transport by low concentrations of cardiac glyco- sides. J. Clin. Invest. 685 1207-1214, 1981. Inagaki, C., Lindenmayer, G.E. and Schwartz, A.: Effects of sodium and potassium on binding of ouabain to the transport adenosine triphosphatase. J. Biol. Chem. 2495 5135-5140, 1974. January, C.T. and Fozzard, H.A.: The effects of membrane potential, extracellular potassium, and tetrodotoxin on the intracellular sodium ion activity of sheep cardiac muscle. Circ. Res. 545 652- 665, 1984. Johnson, E.A.: The effects of quinidine, procaineamide and pyrilamine on the membrane resting and action potential of guinea pig ventri- cular muscle fibers. J. Pharmacol. Exp. Ther. 1115 237-244, 1956. Johnson, E.A. and McKinnon, M.G.: The differnential effect of quini- dine and pyrilamine on the myocardial action potential at various rates of stimulation. J. Pharmacol. Exp. Ther. 1295 460-468, 1957. Johnson, P.N. and Inesi, G.: The effect of methylxanthines and local anesthetics on fragmented sarcoplasmic reticulum. J. Pharmacol. Exp. Ther. 1695 308-314, 1969. Josephson, I. and Sperelakis, N.: Local anesthetic blockade of Ca2+- mediated action potentials in cardiac muscle. Eur. J. Pharmacol. 495 201-208, 1976. Katz, R.I. and Kopin, I.J.: Electrical field—stimulated release of norepinephrine-H from rat atrium: Effects of ions and drugs. J. Pharmacol. Exp. Ther. 1695 229-236, 1969. Kazazoglou, T., Renaud, J-F., Rossi, B. and Lazdunski, M.: Two classes of ouabain receptors in chick ventricular cardiac cells and their relation to (Na+,K+)-ATPase inhibition, intracellular sodium accumulation, Ca2+ influx, and cardiotonic effect. J. Biol. Chem. 2585 12163-12170, 1983. Kim, D-H., Akera, T. and Brody, T.M.: Effects of quinidine on the cardiac glycoside sensitivity of guinea pig and rat heart. J. Pharmacol. Exp. Ther. 2115 559-565, 1981a. Kim, D-H., Akera, T. and Brody, T.M.: Interactions between quinidine and cardiac glycosides involving mutual binding sites in the guinea pig. J. Pharmacol. Exp. Ther. 2185 108-114, 1981b. 125 Kim, D-H., Akera, T. and Brody, T.M.: Tissue-binding sites involved in quinidine-cardiac glycoside interactions. J. Pharmacol. Exp. Ther. 2185 357-362, 1981c. ' Klein, R.L., Holland, W.C. and Tinsley, B.: Quinidine and unidirec- tional cation fluxes in atria. Circ. Res. 85246-252, 1960. Ku, D., Akera, T., Pew, C.L. and Brody, T.M.: Cardiac glycosides: Correlations among Na+,K+-ATPase, sodium pump and contractility in the guinea pig heart. Naunyn-Schmiedeberg's Arch. Pharmacol. 2855 185-200, 1974. Ku, D.D., Akera, T., Tobin, T. and Brody, T.M.: Comparative studies on the effect of monovalent cations and ouabain on cardiac Na+,K+- adenosine triphosphatase and contractile force. J. Pharmacol. Exp. Ther. 1975 458-469, 1976. Langer, G.A.: Sodium exchange in dog ventricular muscle: Relation to frequency of contraction and its role in the control of myocardial contractility. J. Gen. Physiol. 595 1221—1239, 1967. Langer, G.A.: Ionic movements and the control of contraction. In The Mammalian Myocardium, G.A. Langer and A.J. Brady (eds.), 1974, John Wiley and Sons, New York, pp. 193-217. Lash, R.E., Reeves, R.C., Reeves, D.N.S. and Hefner, L.L.: Mechanism of additive effects of digoxin and quinidine on contractility in isolated cardiac muscle. Am. J. Cardiol. 595 483-487, 1982. Leahey, E.B. Jr., Bigger, J.T. Jr., Butler, V.P. Jr., Reiffel, J.A., O'Connell, G.C., Scaffidi, L.E. and Rottman, R.W.: Quinidine- digoxin interaction: Time course and pharmacokinetics. Am. J. Cardiol. 485 1141-1146, 1981. Leahey, E.B. Jr., Hougen, T.J., Bigger, J.T. Jr. and Smith, T.W.: Effect of quinidine on digoxin-inhibited monovalent cation active transport. Circulation 62(Supp1. III): 258, 1980. Leahey, E.B. Jr., Reiffel, J.A., Drusin, R.E., Heisenbuttel, R.H., Lovejoy, W.P. and Bigger, J.T. Jr.: Interaction between quini- dine and digoxin. J. Am. Med. Assoc. 2495 533-534, 1978. Leahey, E.B. Jr., Reiffel, J.A., Giardina, E-G.V. and Bigger, J.T. Jr.: The effect of quinidine and other oral antiarrhythmic drugs on serum digoxin. Ann. Int. Med. 925 605-608, 1980b. Leahey, E.B. Jr., Reiffel, J.A., Heissenbuttel, R.H., Drusin, R.E., Lee, Lee, Lee, Lee, Lee, Lee, Lee, 126 Lovejoy, W.P. and Bigger, J.T. Jr.: Enhanced cardiac effect of digoxin during quinidine treatment. Arch. Int. Med. 1395 519-521, 1979. C.O., and Dagostino, M.: Effect of strophanthidin on intracellu- lar Na ion activity and twitch tension of constantly driven canine cardiac Purkinje fibers. Biophys. J. 495 185-198, 1982. C.O., Kang, D.H., Sokol, J.H. and Lee, K.S.: Relation between intracellular Na ion activity and tension of sheep cardiac Pur- kinje fibers exposed to dihydro-ouabain. Biophys. J. 295 315-330, 1980. K.S., Hume, J.R., Giles, W. and Brown, A.M.: Sodium current depression by lidocaine and quinidine in isolated ventricular cells. Nature 2915 325-327, 1981. K.S. and Klaus, W.: The subcellular basis for the mechanism of inotropic action of cardiac glycosides. Pharmacol. Rev. 235 193- 261, 1971. K.S. and Tsien, R.W.: Reversal of current through calcium channels in dialysed single heart cells. Nature 2925 498-501, 1982. K.S., Weeks, T.A., Kao, R.L., Akaike, N. and Brown, A.M.: Sodium current in single heart mUscle cells. Nature 2285 269-271, 1979. S.G.T.: Quinidine-digoxin interaction. New Engl. J. Med. 3925 175, 1980. Linden, J. and Brooker, G.: Sodium requirement for effects of ouabain on contraction of isolated guinea pig atria. Circ. Res. 465 553— 564, 1980. Lindenmayer, G.E., Laughter, A.H. and Schwartz, A.: Incorporation of inorganic phosphate-32 into a Na+,K+-ATPase preparation: Stimula- tion by ouabain. Arch. Biochem. Biophys. 1215 187-192, 1968. Lindenmayer, G.E. and Schwartz, A.: Nature of the transport adenosine triphosphatase digitalis complex. J. Biol. Chem. 2485 1291-1300, 1973. Lowry, K., Rao, S.N., Pitts, B.J.R. and Askari, A.: Effects of quini— dine on some reactions and ion translocations catalyzed by the Na+,KT—ATPase complex. Biochem. Pharmacol. 225 1369-1377, 1973. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J.: Protein measurement with the Folin phenol reagent. J. Biol. Chem. 1935 265-275, 1951. 127 Matsuda, H. and Noma, A.: Isolation of calcium current and its sensi- tivity to monovalent cations in dialysed ventricular cells of guinea-pig. J. Physiol. 3525 553-573, 1984. — Matsui, H. and Schwartz, A.: Kinetic analysis of ouabain-K+ and Na+ interaction on a Na+,K+-dependent adenosinetriphosphatase from cardiac tissue. Biochem. Biophys. Res. Commun. 255 147-152, 1966. Matsui, H. and Schwartz, A.: Mechanism of cardiac glycoside inhibi- tion of the (Na+-K+)-dependent ATPase from cardiac tissue. Biochim. Biophys. Acta 1515 655-663, 1968. Meier, C.F., Lasseter, K.C. and Pressman, B.C.: Alteration of intra- cellular sodium activity in canine Purkinje fibers by the iono- phore monensin. Pharmacologist 185 122, 1976. Mendez, C., Gruhzit, C.C. and Moe, G.K.: Influence of cycle length upon refractory period of auricles, ventricles, and A-V node in the dog. Am. J. Physiol. 1845 287-295, 1956. Mentrard, D., Vassort, G. and Fischmeister, R.: Changes in external Na induce a membrane current related to the Na-Ca exchange in cesium-loaded frog heart cells. J. Gen. Physiol. 895 201-220, 1984. ~ Mirro, N.J., Watanabe, A.M. and Bailey, J.C.: Electrophysiological effects of disopyramide and quinidine on guinea pig atria and canine cardiac Purkinje fibers. Circ. Res. 465 660-668, 1980. Moran, N.C.: Contraction dependency of the positive inotropic action of cardiac glycosides. Circ. Res. 215 727-740, 1967. Morgan, J.P. and Blinks, J.R.z Intracellular Ca2+ transients in the cat papillary muscle. Can. J. Physiol. Pharmacol. 695 524-528, 1982. Mullins, L.J.: The generation of electric currents in cardiac fibers by Na/Ca exchange. Am. J. Physiol. 2365 C103-C110, 1979. Narahashi, T. and Seyama, I.: Mechanism of nerve membrane depolari- zation caused by grayanotoxin I. J. Physiol. 2425 471-487, 1974. Nash-Alder, P., Louis, C.F., Fudyma, G. and Katz, A.M.: The modifica- tion of unidirectional calcium fluxes by dibucaine in sarcoplasmic reticulum vesicles from rabbit fast skeletal muscle. Mol. Phar- macol. 115 61-65, 1980. Nawrath, H.: Action potential, membrane currents and force of con- traction in mammalian heart muscle fibers treated with quinidine. J. Pharmacol. Exp. Ther. 2165 176—182, 1981. Pang, D.C. and Sperelakis, N.: Veratridine stimulation of calcium uptake by chick embryonic heart cells in culture. J. Mol. Cell. Cardiol. 145 703-709, 1982. 128 Park, M.K. and Vincenzi, F.F.: Rate of onset of cardiotonic steroid- induced inotropism: Influence of temperature and beat interval. J. Pharmacol. Exp. Ther. 1955 140-150, 1975. Pasmooij, J.H., Van Enst, G.C., Bouman, L.M., Allessie, M.A. and Bonke, F.I.M.: The effect of heart rate on the membrane respon- siveness of rabbit atrial muscle. Pflugers Arch. 3665 223-231, 1976. Pedersen, K.E., Christiansen, B.D., Klitgaard, N.A. and Nielsen-Kudsk, F.: Changes in steady state digoxin pharmacokinetics in cardiac patients: Influence of plasma quinidine concentration. Acta Pharmacol. Toxicol. 525 357-363, 1983. Perrone, J.R. and Blostein, R.: Asymmetric interaction of inside-out and right-side-out erythrocyte membrane vesicles with ouabain. Biochim. Biophys. Acta 2965 680-689, 1973. Post, R.L., Hegyvary and Kume, 3.: Activation by adenosine triphos- phate in the phosphorylation kinetics of sodium and potassium ion transport adenosine triphosphatase. J. Biol. Chem. 2425 6530- 6540, 1972. Post, R.L., Merritt, C.R., Kinsolving, C.R. and Albright, C.D.: Membrane adenosine triphosphatase as a participant in the active transport of sodium and potassium in the human erythrocyte. J. Biol. Chem. 2355 1796-1802, 1960. Post, R.L., Sen, A.K. and Rosenthal, A.S.: A phosphorylated inter- mediate in the ATPdependent sodium and potassium transport across kidney membranes. J. Biol. Chem. 2495 1437-1445, 1965. Powell, R., Fenster, P., Wandell, M., Hager, D., Graves, P., Conrad, K. and Goldman, S.: Quinidine-digoxin interaction: Multiple-dose pharmacokinetics. Clin. Pharmacol. Ther. 275 279, 1980. Prindle, K.H. Jr., Skelton, C.L., Epstein, S.E. and Marcus, F.I.: Influence of extracellular potassium concentration on myocardial uptake and inotropic effect of tritiated digoxin. Circ. Res. 285 337-345, 1971. ' Pruett, J.K. and Woods, E.F.: The relationship of intracellular de- polarization rates and contractility in the dog ventricle in_situ: Effects of positive and negative inotropic agents. J. Pharmacol. Exp. Ther. 1575 1-7, 1967. Repke, K.: Metabolism of cardiac glycosides. In Proceedings of the lst International Pharmacology Meeting, vol. 3, pp. 47-73. Pergamon Press, Oxford, 1963. 129 Reiter, H. and Seitz, N.: The dependence of calcium efflux from car- diac muscle on temperature and external ion composition. J. Physiol. 1955 451-470, 1968. Reuter, H. and Scholz, H.: A study of the ion selectivity and the kinetic properties of the calcium-dependent slow inward current in mammalian cardiac muscle. J. Physiol. 2545 17-47, 1977. Risler, T., Peters, U., Grabensee, B. and Seipel, L.: Quinidine- digoxin interaction. New Engl. J. Med. 3925 175, 1980. Robinson, J.D. and Flashner, M.S.: ‘The (Na++K+)-activated ATPase enzymatic and transport properties. Biochim. Biophys. Acta 5495 145-176, 1979. Sanchez-Chapula, J., Tsuda, Y. and Josephson, J.R.: Voltage- and use— dependent effects of lidocaine on sodium current in rat single ventricular cells. Circ. Res. 525 557-565, 1983. Schatzmann, H.J.: Herzglycoside als Hemmstoffe fur den aktiven Kalium- und Natrium-transport durch die Erythrocytenmembran. Helv. Physiol. Acta 115 346-354, 1957. Schenck-Gustafsson, K., Jogestrand, T., Norlander, R. and Dalqvist, R.: Effect of quinidine on digoxin concentration in skeletal muscle and serum in patients with atrial fibrillation. New Engl. J. Med. 3955 209-211, 1981. Schwartz, A., Allen, J.C., Van Winkle, W.B. and Munson, R.: Further studies on the correlation between the inotropic action of ouabain and its interaction with the Na+,K+-adenosine triphsophatase: Isolated perfused rabbit and cat hearts. J. Pharmacol. Exp. Ther. 1915 119-127, 1974. Schwartz, A., Matsui, H. and Laughter, A.H.: Tritiated digoxin binding to (Na++K+)-activated adenosine triphosphatase: Possible allosteric site. Science 1605 323-325, 1968. Schwarz, W., Palade, P.T. and Hille, B.: Local anesthetics: Effect of pH on use-dependent block of sodium channels in frog muscle. Biophys. J. 295 343-368, 1977. Sen, A.K., Tobin, T. and Post, R.L.: A cycle for ouabain inhibition of sodium- and potassium—dependent adenosine triphosphatase. J. Biol. Chem. 2445 6596-6604, 1969. Siegel, G.J. and Josephson, L.: Ouabain reaction with microsomal (sodium-p1us-potassium)-activated adenosinetriphosphatase. Eur. J. Biochem. 255 323-335, 1972. Simons, T.J.B.: Potassiumzpotassium exchange catalysed by the sodium pump in human red cells. J. Physiol. 2315 123-155, 1974. 130 Sjodin, R.A.: Rubidium and cesium fluxes as related to the membrane potential. J. Gen. Physiol. 425 983-1003, 1959. Skou, J.C.: The influence of some cations on an adenosine triphos- phatase from peripheral nerves. Biochim. Biophys. Acta 235 394- 401, 1957. Skou, J.C. and Hilberg, C.: The effegE of cations, g- strophanthin and oligomycin on the labeling from [ P]ATP of the (Na ++K+)- activated enzyme system and the effect of cations and g- -strophanthin on the labelling from [32P]ITP and 32Pi’ Biochim. Biophys. Acta l§5_: l98- 219, 1969. Sperelakis, N.: Origin of the cardiac resting potential. In Handbook of Physiology, sec. 2, The Cardiovascular System, ed. by R.M. Berne, vol. 1, PP. 187-267, Waverly Press, Baltimore, 1979. Steiness, E., Waldorff, S., Hansen, P.B., Kjaergard, H., Buch, J. and Egeblad, H.: Reduction of digoxin-induced inotropism during quinidine administration. Clin. Pharmacol. Ther. 225 791-795, 1980. Straub, K.D., Kane, 3.3., Bissett, J.K. and Doherty, J.E.: Alteration of digitalis binding by quinidine: A mechanism of digitalis- quinidine interaction. Circulation 57/58(Suppl. II): 58, 1978. Strichartz, G. and Cohen, I.: Vm x as a measure of GNa in nerve and cardiac membranes. Biophys. 3. 23: 153- 156, 1978. Temma, K. , and Akera, T. Enhancement of cardiac actions of ouabain and its binding to Na+ ,K+ -adenosine triphosphatase by increased sodium influx in isolated guinea- pig heart. J. Pharmacol. Exp. Ther. 2235 490-496, 1983. Temma, K. and Akera, T. Decreases in active sodium pumping sites and their interaction With ouabain caused by low Na+ incubation of isolated guinea- pig atrial muscle. J. Pharmacol. Exp. Ther. 2255 660-666, 1983. Tobin, T. ,Baskin, S. I. ,Akera, L and Brody, T. M. Nucleotide speci- ficity of the Na+ —stimu1ated phosphorylation and [3 H]0uabain- binding reactions of (Na++K+ )- -dependent adenosine triphosphatase. Mol. Pharmacol. §_: 256- 263, 1972. Undrovinas, A. I. ,Yushmanova, A. Y. ,Hering, S. and Rosenshtraukh, L. V. Voltage clamp method on single cardiac cells from adult rat heart Experientia 36. 572—574,1980. van Zwieten, P.A.: Decrease in ionic permeability of the cell mem- brane in guinea-pig atrial tissue by treatment with anggfibrilla- tory agents and hexobarbitone, determined by means of Brit. J. Pharmacol. 355 103-111, 1969. 131 Vaughan-Williams, E.M.: The mode of action of quinidine on isolated rabbit atria interpreted from intracellular records. Brit. J. Pharmacol.1§5 276-287, 1958. Walton, M. and Fozzard, H. The relation of Vma Xto INa’ GNa’ and hco in a model of the cardiac Purkinje fiber. mBiophys. J. 25: 407- 420,1979. Warner, N.J., Barnard, J.T., Leahey, E.B. Jr., Hougen, T.J., Bigger, J.T. Jr., and Smith, T.W.: Myocardial monovalent cation trans- port during quinidine—digoxin interaction in dogs. Circ. Res. 545 453-460, 1984. Wassermann, O. and Holland, W.C.: Effects of tetrodotoxin and ouabain on atrial contractions. Pharmacol. Res. Commun. 15 236-241, 1969. Wasserstrom, J.A., Schwartz, D.J. and Fozzard, H.A.: Relation between intracellular sodium and twitch tension in sheep cardiac Purkinje strands exposed to cardiac glycosides. Circ. Res. 525 697-705, 1983. Weidmann, S.: The effect of the cardiac membrane potential on the rapid availability of the sodium-carrying system. J. Physiol. 1215 213-224, 1955. Weld, F.M., Coromilas, J., Rottman, J.N. and Bigger, J.T., Jr.: Mecha- nisms of quinidine-induced depression of maximum upstroke velocity in ovine cardiac Purkinje fibers. Circ. Res. 595 369-376, 1982. Wiggins, J.R. and Bentolila, J.J.: Sodium dependence of the positive inotropic effect of cardiac glycosides. J. Pharmacol. Exp. Ther. 2155 569—574, 1980. Wilkerson, R.D., Beck, B.L. and Orlowski, E.W.: Quinidine increases the serum concentration of digoxin but masks increased cardiac effects. Fed. Proc. 435 1044, 1984. Williams, J.F. Jr. and Mathew, B.: Effect of quinidine on positive inotropic action of digoxin. Am. J. Cardiol. 415 1052-1055, 1981. Yamamoto, S. ,Akera, T. and Brody, T. M. Sodium influx rate and ouabain- sensitive rubidium uptake in isolated guinea pig atria. Biochim. Biophys. Acta 555: 270— 284, 1979. Zaman, L., Pastore, J.D. and Kosowsky, 8.0.: The digitalis-quinidine interaction: Echocardiographic evidence of enhanced left ventri- cular contractility in man. Clin. Res. 295 253A, 1981. iiiii HIWMM'M’((‘3111'111111111s