“ﬁt-.4218

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

3H-Ouabain Binding and Sodium-Pump Activity Measured
in Myocytes Isolated from Guinea—pig Heart

presented by

Paul Stemmer

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degree in Pharmacology

mm WW4

Major professor ‘

7/27/86

Date

um”... awn—“- ‘ ' I" ”f ' ' 1 ’ 0-127"

 

 

MSU I

LIBRARIES

 

”-

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

3HmOUABAIN BINDING AND SODIUM-PUMP ACTIVITY MEASURED

IN MYOCYTES ISOLATED FROM GUINEA-PIG HEART

BY

Paul Stemmer

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
1986

ABSTRACT

3H—Ouabain Binding and Sodium-Pump Activity Measured in
Myocy tes Isolated from Guinea-Pig Heart
by

Paul Stemmer

2+

Myocytes, quiescent in millimolar concentrations of Ca , were isolated

from guinea-pig heart by treatment with collagenase and hyaluronidase; greater
than 8096 of cells were rod-shaped. One micromolar ouabain protected isolated
myocyte preparations from loss of viable cells during a 60-min incubation. One
millimolar ouabain was toxic to cells within 60 min as determined by loss of the
rod shape. Because of toxicity of millimolar ouabain, non-specific 3H—ouabain

binding was assessed by monitoring dissociation of bound drug. Analysis of

3

specific H—ouabain binding to myocytes yielded non-linear Scatchard plots. Non-

3

linearity appears to result from reduced H-ouabain binding due to low intracellu-

lar Na+ concentration. Addition of 2 uM monensin, a Na+ ionophore, significantly

increased 3H-ouabain binding. Incubation in Ca2

3

+-free solution (0.25 mM EGTA)
stimulated H—ouabain binding to a greater degree than monensin and caused
Scatchard plots to have two distinct linear components. Monensin had no
significant effects when 3H—ouabain binding occurred in Carr-free solution.
Effects of Ca2+-free incubation to increase 3H-ouabain binding suggest that Ca2+
has a direct effect on 3H-ouabain binding. Alternatively, Ca2+-free incubation

may increase Na+ permeability of the sarcolemma. Isoproterenol, phenylephrine,

Paul Stemmer

4.

TPA (phorbol 12-myristate 13-acetate), La3 , and the Ca2+-ionophore A2318?

failed to cause significant changes in 3IrI-ouabain binding when myocytes were

3 2

incubated in a solution containing 0.5 or 2.5 uM H—ouabain, 0.1 mM Ca + and 1

mM K+. Caffeine decreased binding of 3I-l-ouabain and the percentage of rod-

shaped cells. Activity of the sodium—pump was estimated in myocytes using

86

ouabain-sensitive Rb+ up take. Monensin or Na+-loading in a K+-free, Rb+-free

86Rb+ uptake by myocytes, indicating

solution caused up to a four-fold increase in
enhanced Na+-pump activity. Stimulation of pump activity by monensin peaked at
30 11M. Stimulation by Na+-loading plateaued after a 45-min incubation for Nat
loading at 37°C. Under the above conditions, intracellular Na+ is apparently no
longer rate-limiting for turnover of the sodium—pump. In addition, these interven-
tions caused concentra don-response curves for ouabain-induced Na+-pump inhibi-

6

tion to be shifted to the left. The presence of Ca2+ during 8 Rb+ uptake reduced

this measure of sodium-pump activity in non-treated and stimulated myocytes

2

indicating that extracellular Ca + reduces the Na+ load to the sodium-pump or

acts directly to inhibit pump activity.

ACKNOWLEDG EM ENTS

I sincerely thank Dr. Tai Akera for his invaluable guidance during my
graduate training. His patience, support, criticism and advice have been
instrumental in my development as a scientist. Most of all I thank him for the
confidence and trust he has shown in my work and in me.

Isincerely appreciate the advice and criticisms offered during the course of
my graduate study by Drs. Theodore M. Brody, Richard H. Kennedy, Yuk-Chow Ng
and Ernst Seifen. I also thank the other members of my guidance committee.
Drs. William W. Wells and Joseph R. Hume each have been supportive and have
contributed significantly to my understanding of the work I have done.

I deeply appreciate the help given to me by Mrs. Eileen Allison while I have
been at Michigan State. Iam also extremely grateful to Ms. Cynthia Clingan, Ai-
zhen Yao and Isabel Corcos. These diree have worked diligently and conscien-
tiously in performing the experiments detailed in this text. They are excellent
coworkers and Isincerely thank each of them.

I thank my classmates, Patti Ganey and Nancy Shannon for their friendship
and help. I also thank Drs. Kyosuke Temma, Yumi Katano, Yuji Nirasawa, Josh
Berlin and Dong-Hee Kim for their friendship, encouragement and advice. I also
thank Ms. Diane Hummel for her assistance throughout my graduate training and
especially during preparation of this manuscript.

Finally, I thank my parents and Roseann Vorce for their confidence and

support which have been invaluable to me.

ii

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

INTRODUCTION
A. General background
B. Factors affecting glycoside binding to Na,K-ATPase
C. Factors affecting sodium-pump activity
D. Isolated myocytes as a preparation for H—ouabain binding
and ion flux studies
E. Objectives

M ATERIALS AND METHODS

A.
B.
C.
D.
E.

F.
G.
H.
I.
RESULTS

A.

Myocyte isolation

H-Ouabain equ' ibrium binding studies
gassogiation of H-ouabain from myocytes

Rb uptake studies

Separation of myocytes and bound ligands from non-bound
ligands
Silane treatment fogfiberglass filters
Experiments with K -deficient animals
Miscellaneous methods
Materials

Isolation of myocytes from guinea-pig heart

1. Viability of guinea-pig myocytes incubated in the pre-
sence or absence of ouabain

Binding of 3H—ouabain to myocytes isolated from guinea-pig
heart

1. Estimation of non-specific 3H-ouabain binding to
guinea—pig myocytes 3

2. Estimation of kinetic parameters for H—ouabain bind-
ing to guinea-pig myogytes

3. Extracellular K and H-ouabain binding

iii

10
11

12

17
18

20
23
24
25
26

26

29

30

33

42
56

TABLE OF CONTENTS (continued)

4. Binding of 3H—ouabain to myocyteszin the presence of
agents which alter intracellular Ca concentration or
stimulate production of second messengers

C. Sodium-pump activity and capacity assessed in myocytes
1. Activity and capacity of the sodium-pump
2. Reduction of reservp capacity by interventions increas-

ing intracellular Na concentration or Na influx
DISCUSSION

A. Regulation of the Na,K-ATPase and cardiac glycoside binding
in intact tissue 3

B. Myocyte isolation for examination of H—ouabain binding and
sodium-pump agtivity

C. Estimation of H-ouabain binding to guinea-pig cardiac myo-
cytes

D. Estimation of sodium—pump activity and capacity in guinea-

pig cardiac myocy tes

SUMMARY AND CONCLUSIONS

BIBLIOGRAPHY

iv

Page

66
73
75

88

92

92
93
96
117

125

Table

LIST OF TABLES

Estimates of 3H-ouabain binding to myocytes

Binding and dissociation o; 3H-ouabain to guinea-pig myo-
cytes incubated in 2.5 uM H—ouabain

3H-ouabain binding to guinea-pig myocyteﬁin the presence
of agents known to affect intracellular Ca concentration
or stimulate production of second messengers

39

63

69

Figre

10

11

12

13

LIST OF FIGURES

Photograph of ventricular myocytes isolated from guinea-
pig heart ‘
Elecron micrograph of a single ventricular myocyte from
guinea-pig heart

Viability of myocytes from guinea-pig heart in the presence
of ouabain

Photograph of myocytes from guinea-pig heart following a
60-min incubation with 1 mM ouabain

Dissociation of 3H-ouabain from myocy tes - time course

Dissociation of 3H—ouabain from myocytes - estimation of
k

Binding of 3H—ouabain to rapidly and slowly dissociating
sites

Estimation of kinetic parameters for 3
the displacement method

Time course of 3
dissociating sites

H—ouabain binding by
H-ouabain binding to the rapidly and slowly
Concentration dependence for 3H—ouabain binding to myo-

cytes, 1-20 11M H-ouabain

Concentration depfndence for 3H-ouabain binding to myo-
cytes, 0.5-3.0 uM H-ouabain

Specific 3H-ouabain binding to guinea-pig myocytes, 1-20
11M H—ouabain

Specific33H-ouabain binding to guinea-pig myocytes, 0.05-
3.0 uM H-ouabain

vi

27

28

31

34
36

37

41

43

45

48

49

51

52

LIST OF FIGURES (Continued)

Figre
14

15

16
17
18

19
20

21

22

23

24

25

26

Scatchard pigts describing 3

sence of Ca

H—ouabain binding in the pre-
Scatchard p296 describing 3H—ouabain binding in the ab-
sence of Ca

Scatchard plots describing 3H—ouabain binding

Dissociation of 3H—ouabain from myocytes - kinetic analysis

Redistribution of K+ from myocytes in the presence of
ouabain

Monensin stimulates sodium-pump activity in myocytes
Capacity of the sodium-pump

Ouabain-sensitive 86Rb+ uptake is inhibited by Ca
concentration-dependent manner

+.
21na

Stimulatiop of sodium-pump activity by incubation of myo—
cytes in K -free, Rb -free solution

Sodium-pump activity is not increased in myocytes from K+-
deficient animals

Inhibition of sodium-pump activity by ouabain: Effect of
monensin and Ca

Concentration-response curves for ouabain inhibition of so-
dium—pump activity: Effects of monensin and Ca

Scatchard plots describing models for positive cooperativity
in binding

vii

54

55
57
61

65
77

79

81

84

87

89

90

112

INTRODUCTION

A. General Background

 

It is generally accepted that Na,K-ATPase is the pharmacological receptor
for the therapeutic and toxic effects of cardiac glycosides (Lee and Klaus, 1971;
Akera and Brody, 1978, 1982). The positive inotropic effect of these agents is
believed to result from inhibition of the Na,K-ATPase. Inhibition of the Na,K-
ATPase causes Na+ transients associated with membrane depolarization to be
larger in magnitude and of longer duration (Bentfeld e_t Q” 1977; Akera and
Brody, 1978). Also associated with Na,K—ATPase inhibition is an increase in
resting intracellular Na+ concentration (Lee _e_t g” 1980; Lee and Dagostino,
1982). Changes in the sodium transient and resting intracellular Na+
concentration affect force of contraction by altering the dynamic and equilibrium

2

activity of the Na/Ca exchange mechanism which in turn affects Ca + transients

2+ stores (Akera and Brody, 1985; Lee e_t a_l.,

and the magnitude of intracellular Ca
1985). Minor changes in intracellular Na+ concentration are able to cause
significant changes in the transport of Ca2+ via the Na/Ca exchange mechanism
because this system is extremely sensitive to changes in the Na+ gradient caused
by changes in intracellular Na+ concentration (Mullins, 1979; Langer, 1982).
Larger changes in the Na)r transient and intracellular Na+ concentration which are
associated with 60 to 80% inhibition of the Na,K-ATPase (Akera and Brody, 1978)
result in "Ca2+ overload" which causes toxic effects of the cardiac glycosides.

Although an overwhelming amount of evidence supports the theory that all

direct effects of cardiac glycosides on the heart are the result of Na,K-ATPase

inhibition, other theories have been proposed to explain the inotropic effects of
low concentrations of cardiac glycosides. These are that glycoside binding to the
Na,K-ATPase affects Ca2+ binding to the intracellular side of the sarcolemmal
membrane (Schwartz, 1976; Gervais gt a_1., 1977; Lullmann and Peters, 1979;
Preuner, 1979; Schwartz and Adams, 1980) or that binding to a different receptor
or different Site on the Na,K-ATPase than the one which causes enzyme inhibition
results in a positive inotropic effect which is associated with a stimulation of the
Na,K-ATPase (Ghysel-Burton and Godfraind, 1979; Godfraind and Ghysel—Burton,
1980; Noble, 1980; Godfraind gt a_1., 1982). Decreases in tissue Na+ content and
increases in tissue K+ content of cardiac muscle exposed to low concentrations of
cardiac glycosides have been offered as evidence for stimulation of the Na,K-
ATPase (Godfraind and Ghysel—Burton, 1977; Ghysel-Burton and Godfraind, 1979);
however, a mechanism by which stimulation of the Na,K—ATPase could produce or

+ O I
2 concentration 13 unknown.

be caused by an increase in the intracellular Ca
Therefore, the theory is not generally accepted. High affinity binding sites other
than the Na,K-ATPase for the cardiac glycosides have not been found in
homogenates of cardiac muscle. This, however, could be due to some factor in
intact tissue which is essential for binding of these agents to a site other than the
Na,K—ATPase being lost during tissue homogenization.

In order to unequivocally confirm that all direct actions of cardiac glyco-
sides are due to the binding to and inhibition of Na,K-ATPase, glycoside binding
must be studied in intact tissue or intact cardiac muscle cells. Binding must then
be shown to correlate with inhibition of the sodium-pump. The sodium-pump is
the physiological representation of the Na,K-ATPase. The goal of this project
was to characterize binding of one cardiac glycoside, 3H-ouabain, to Ca2+

tolerant myocytes from guinea-pig heart and then to further determine if

intracellular Ca2+ concentration or phosphorylation of membrane constituents can

affect either glycoside binding or sodium-pump function in a manner which could
affect the sensitivity of the heart to therapeutic or toxic effects of the cardiac

glycosides.

B. Factors Affecting£lycoside Binding to Na,K-ATPase

Binding of cardiac glycosides is believed to be a second order reaction which
can be considered to be a pseudo first-order reaction, described kinetically as a
first-order reaction, when an excess of glycoside is present in the reaction
mixture (Gelbart and Goldman, 1977). Binding of the glycosides is dependent on
the conformation of the Na,K—ATPase (Akera e_t a_1., 1976). Enzyme conformation
changes as the Na,K—ATPase binds different ligands and promotes the transport of
ions. The reaction cycle of Na,K-ATPase is considered to be represented by the

following scheme (Akera and Brody, 1978; Yoda and Yoda, 1986).

E +ATP—->E ~P+ADP

\ translocation of Na+

P

. ATP + Nat—2+9 News

1 Mg

1 1

Na 'E1
on \
K ATP , EP-Ou
A ??
Na‘E*P

 

Pi Ou
K‘ ale—— K'EZP <— K+ + EzP————> EzP-Ou

translocation of K+ K1

 

K ' EZP-Ou
Ouabain binds to phosphorylated intermediates of the enzyme particularly the EZP

3H-ouabain

intermediate (Matsui and Schwartz, 1968). It has not been shown that
binds to intermediates other than the E2P form of the Na,K-ATPase, however,
curved Scatchard plots describing 3H-ouabain binding to isolated Na,K-ATPase

(Hansen, 1976; Wellsmith and Lindenmayer, 1980; Godfraind e_t a_1., 1980) indicate

that this ligand is binding to more than a single class of binding Site. Dependent
on the presence or absence of K+ during binding, the Scatchard plots describing
the binding become linear (Hansen, 1976; Godfraind e_t 31:, 1980). Hansen (1976)
showed that addition of 2 mM K+ caused Scatchard plots to become linear when
the 3H--ouabain binding to Na,K-ATPase occurred in a Mg2+ plus phosphate
system. Godfraind and coworkers (1980) showed that in an incubation mixture
containing Na+, Mg2+, ATP, phosphate and vanadate, linear Scatchard plots were

3

obtained when H—ouabain binding occurred in the absence but not in the presence

of K+. The ability of K+ to affect 3

H-ouabain binding under each incubation
condition suggests that these sites are interconvertible and represent different
conformations of the Na,K-ATPase. Wellsmith and Lindenmayer (1980) found the
3H-ouabain binding sites in their preparation were not interconvertible, but that
binding to each Site was affected by the presence of K+. Therefore, these
investigators propose that one of the binding sites is an inactive form of Na,K-
ATPase. Other evidence indicating 3H—ouabain can be bound to different

3

conformations of Na,K-ATPase is that dissociation of H—ouabain from isolated

enzyme is described by two exponential equations. These equations are believed

3H-ouabain from different forms of the Na,K—ATPase.

to describe dissociation of
The relative abundance of each conformation of the enzyme is determined by the
presence of K+ (Choi and Akera, 1977; Godfraind e_t g” 1980).

Ligand environment is the ultimate determinant of the kinetics of 3H-
ouabain binding. Glycoside binding is supported by ligands which promote
conformation changes of the enzyme to a binding conformation (Matsui and
Schwartz, 1968; Akera g_t_ gs 1974). These ligands are Na+, Mg2+ and ATP
together or Mg2+ and inorganic phosphate together. Other combinations of
ligands, such as Mg2+ and ATP, can support glycoside binding to a limited extent

by affecting enzyme conformation. Alternatively, glycoside binding is inhibited

by ligands which promote changes which take the enzyme out of a binding
conformation (Akera and Brody, 1970; Tobin e_t gl_., 1974; Inagaki e_t a_1., 1974).
The species which do this are K+ itself, or ions such as Rb+ or Li+ which
substitute for K+ and promote dephosphorylation of the E2P intermediate.

2

Inhibition of Na,K—ATPase activity by Ca + has been shown to occur and be

caused by competitive inhibition of Na+ binding (Tobin gt at, 1973). This
inhibition requires high concentrations of Ca2+ (IC50 = 0.5 mM) relative to what is
known to be present in cardiac muscle where maximum interaction of contractile

2+ reaches a concentration of 10 11M (Fabiato and

2

filaments occurs when Ca
Fabiato, 1970; Solaro gt gt, 1974). Therefore, Ca + itself is not considered to
have direct effects on Na,K-ATPase.

In intact cells, affinity of Na,K-ATPase for cardiac glycosides is also
affected by the ligand environment. The same principles which apply to the
isolated enzyme apply to the Na,K-ATPase in intact cells. Those are, ligands
which cause changes to a binding conformation (e.g., intracellular Na+) stimulate
glycoside binding and those which promote changes out of a binding conformation
(e.g., extracellular K+) inhibit glycoside binding (Akera and Brody, 1971; Yama-
moto gt g” 1980; Temma and Akera, 1982; Kennedy gt a_1., 1983). It is possible
however, that in intact cells other mechanisms exist which can affect the affinity
of the enzyme for cardiac glycosides or the activity of the sodium-pump at given
Na+ and K+ concentrations. If such mechanisms exist, then their actions could
have significant effects on the therapeutic and/or toxic actions of the cardiac
glycosides.

One factor, which is intrinsically tied to the positive inotropic effect of
cardiac glycosides and has been reported to affect Na,K-ATPase activity and 3H-
ouabain binding under very specific conditions, is Ca2+. Reports implicating Ca2+

as a factor involved in modulation of the sodium-pump or cardiac glycoside

binding are not supported by experiments examining the effect of Ca2+ on

2

isolated Na,K-ATPase. Therefore, another factor, such as a Ca +-binding protein,

2+ on the Na,K-ATPase which occurs at

2+

is likely to be involved in any effect of Ca

2

+ concentrations. The potential interaction between Ca and

2

physiological Ca

Na,K-ATPase activity or Ca + and cardiac glycoside binding is very intriguing. If

2+

intracellular Ca stimulates Na,K-ATPase activity, either by itself or via a

Ca2+—binding protein, it would provide a feedback mechanism to control the

2+ transients which affect force of

2+

effects of sodium—pump inhibition on Ca
contraction. Alternatively, if intracellular Ca inhibits glycoside binding, it may
provide a feedback mechanism to control the effects of any endogenous digitalis-
like substances.

Removal of Ca2+ has been reported to increase the affinity of the Na,K—
ATPase from murine plasmocy toma cell membranes for ouabain as determined by
a shif t in the concentration response curve for ouabain inhibition of Na,K-ATPase
activity (Zachowski e_t a_1., 1977; Lelievre e_t a_1., 1979). Also, short exposure (6
min) of rat hearts to Cay-free medium has been reported to increase the affinity
of Na,K-ATPase purified from those hearts for ouabain as determined by the same
assay (Mansier and Lelievre, 1982; Mansier gt a_1., 1983). The increase in affinity,
caused by removal of Ca2+ from murine plasmocytoma cell membranes, may be
due to removal of some Ca2+-binding protein from the membrane. Evidence for
this is that the normal affinity of the Na,K-ATPase for ouabain in these

2+ 2

membranes depleted of Ca + and

2

with EDTA is restored by addition of Ca
Ca +-binding proteins (Charlemagne e_t e_t, 1980; Geny e_t Q” 1982) in a
calmodulin—dependent manner (Lelievre g .91., 1985). The Ca2+-binding proteins

2+ and

which decreased the affinity of this Na,K-ATPase for ouabain in a Ca
calmodulin—dependent manner were tropornyosin (Charlemagne e_t g, 1980; Lelie-

vre gt a_1., 1985) and a B-actinin—like protein purified from the cells (Geny e_t a_1.,

1982). The effective concentration of Ca2+ was less than 1 uM (Lelievre g_t a_1.,
1985). These experiments demonstrate that, in at least one cell type, it is
possible to modulate affinity of the Na,K-ATPase for ouabain in a manner

2+ at concentrations which occur within heart muscle and which

dependent on Ca
do not affect activity of the enzyme.

Isolated Na,K-ATPase from rat hearts exposed to Ca2+-free medium also
has an increased affinity for ouabain (Mansier and Lelievre, 1982; Mansier gt gt,
1983). The significance of this is more difficult to evaluate because rat heart is
known to contain two forms of Na,K-ATPase with different affinities for the
cardiac glycosides (Erdmann gt gl., 1980; Noel and Godfraind, 1984). Therefore, it
is possible that removing Ca2+ somehow caused the high affinity form of the
enzyme to be selectively retained during isolation. This appears to be extremely
unlikely, however, and is seemingly refuted by the observation that preincubation
of the ouabain-sensitive enzyme with 10 mM K+ totally reverses the high affinity
inhibition of Na,K-ATPase activity by ouabain of a small percentage of the
enzyme (Mansier gt g” 1983). Efforts to reverse the effect of Ca2+-free
perfusion to increase affinity of Na,K-ATPase, by incubation with Ca2+ or Ca2+-
binding proteins after the enzyme isolation, were not reported. These studies

2

suggest a link between Ca + and affinity of Na,K-ATPase from cardiac muscle for

cardiac glycosides.

C. Factors Affecting Sodium-me Activity

Activity of the Na,K—ATPase can also be studied with isolated enzyme
preparations or in intact tissue. From experiments done with isolated enzyme it
is known that Na+ and K+ each stimulate sodium-pump activity in a concentra-
tion-dependent manner. The maximum effect of Na+ is not achieved unless the

Na+ concentration is greater than 100 mM (Akera e_t g, 1985) whereas the Km

for K+ activation has been estimated to be between 0.9 and 6.3 mM (Cohen e_t g”

1984).

Experiments examining effects of Ca2+

2

on Na,K-ATPase activity have

shown that Ca + inhibits enzyme activity (Tobin gt gl_., 1973; Godfraind e_t gl_.,

+ O
2 at micromolar

1977; Beauge and Campos, 1983). A single report that Ca
concentrations stimulates activity of Na,K-ATPase in brain homogenates has been
presented (Powis gt gl_., 1983). The latter investigators found that if Ca2+ was
present at a concentration of 0.1 to 1.0 11M in the absence of any chelating
agents, the Na,K-ATPase activity of rat brain homogenates was increased
approximately 2596. The presence of calmodan with 0.1 to 1.0 uM concentra-

tions of Ca2+

caused enzyme activity to increase approximately 5596. III the
presence of EDTA, basal Na,K-ATPase activity was increased 12096 and the
effect of Ca2+ to stimulate the enzyme was eliminated. Higher concentrations of

2

Ca + inhibited enzyme activity with an 1050 of approximately 200 uM, irrespec-

2+ on Na,K-ATPase

tive of the presence of EDTA. This inhibitory effect of Ca
activity is consistent with the work of other investigators who find that the [C50
for inhibition of enzyme activity by Ca2+ is approximtely 0.5 mM (Tobin gt gl.,
1973; Godfraind e_t g” 1977; Beauge and Campos, 1983). The report that low
concentrations of Ca2+ stimulate Na,K-ATPase activity (Powis g_t g, 1983) is
unique but suggests there may be a feedback mechanism other than the Na+
concentration to regulate the Na,K-ATPase. The ability of calmodulin to amplify
the stimulatory effect of Ca2+ suggests that it is mediated by a Ca2+-binding
protein.

An opposite effect of a factor from human red blood cells has been
reported. This factor decreases the 1050 for Ca2+ inhibition of Na,K-ATPase
activity to 1 11M (Yingst and Marcovitz, 1983; Yingst, 1983; Yingst and Polasek,

1985). The nature of the inhibitory factor and the mechanism by which it

2+ to inhibit the Na,K-ATPase are unknown. The presence of

2

interacts with Ca
factors in cardiac muscle which can interact with Ca + to stimulate or inhibit
Na,K-ATPase in that tissue has not been demonstrated. However, their reported
presence in other tissues suggests that similar mechanisms for control of the
sodium-pump may exist in the heart.

Sodium-pump activity is the physiological representation of the Na,K-
ATPase. Measurement of sodium-pump activity can be accomplished in sodium-
loaded cardiac muscle by monitoring a transient outward current upon exposure of
those tissues to KL, Rb+ or Cs+ (Gadsby and Cranefield, 1979; Eisner and Lederer,
1980; Daut and Rudel, 1982). This current is inhibited by cardiac glycosides and,
therefore, is attributed to activity of the Na,K-ATPase.

The more commonly used method to monitor Na,K-ATPase activity in intact
cardiac muscle is by counter-ion transport. This method requires measurement of

42K+ with time in the presence and absence of ouabain.

42

the uptake of 86Rb+ or

86 K+ uptake is believed to represent ion uptake

The ouabain-sensitive Rb)r or
which is dependent on Na,K-ATPase activity.

In intact cardiac muscle, the intracellular Na+ activity is less than 10 mM.
Activation of Na,K-ATPase, however, is not maximal at less than 100 mM Na+.
Therefore, as the Na+ inﬂux rate increases, sodium-pump activity will also
increase so that only modest increases in intracellular Na+ concentration occur
(Akera and Brody, 1985). This feature of the sodium-pump is responsible for the
observed reserve capacity of this enzyme system in cardiac muscle. The fact that
the sodium-pump has a reserve capacity means that the measurement of Na,K-
ATPase activity in intact tissue is really a measure of Na+ influx rate unless Na+
is accumulating in the tissue or has accumulated to the point that the Na+

concentration is not rate-limiting for Na,K-ATPase activation (Yamamoto _e_t gl_.,

1979; Akera e_t a_1., 1981; Akera and Brody, 1985).

10

In order to determine if an intervention has a direct effect on the Na,K-
ATPase an increase or decrease in reserve capacity of the sodium-pump must be
documented. Direct modulation of the Na,K-ATPase, other than by specific
inhibitors, has not been demonstrated in intact tissue although a decrease in
reserve capacity has been Shown to occur when Na+ influx is increased by
electrical stimulation (Yamamoto e_t gl., 1980). The decrease in reserve capacity
is seen as a shif t to the left of the digoxigenin concentration response curve for

86Rb+ uptake. Stimulation of sodium-pump acti-

inhibition of ouabain-sensitive
vity in isolated cardiac myocytes, by interventions causing an increase in Na+
inﬂux or by sodium-loading the cells, has not been demonstrated. Measurement of
sodium-pump activity in Ca2+-tolerant myocytes rather than in intact tissue
offers the advantage of only one cell type being present. Additionally, removal of
all barriers for ion diffusion may allow for a more accurate determination of
sodium-pump activity.

D. Isolated Myocytes as a Preparation for 3

Studies

H-Ouabain Bindigg and Ion Flux

Quantitating cardiac glycoside binding to intact cardiac muscle has been
attempted with only limited success (Busse g_t a_1., 1979; Kjeldsen g_t a_1., 1985;
Herzig gt a_1., 1985b). Problems associated with this are a high degree of non-
specific binding ( 5096 of total binding) and that binding occurs to receptors on
many different cell types.

Analysis of kinetic parameters for 3H—ouabain binding has been successfully
accomplished using data describing binding to myocy tes dissociated from hearts of
adult animals (Onji and Liu, 1981; Adams e_t gl_., 1982) or cultured cells derived
from hearts in the embryonic stage of development (Friedman gt gl_., 1980;
Kazazoglou g gl_., 1983; Kinor e_t gl_., 1984; Werden _e_t g1_., 1984). Each of these

preparations appear to be better suited than intact muscle for experiments where

11

determination of kinetic parameters for drug binding is desired. Each of these
preparations also has a drawback, however. Myocytes from adult animals always
exist as a mixed population of living, dead and dying cells. Therefore, it is

necessary to develop a selection procedure to isolate viable cells with good

stability in the presence of millimolar concentrations of Ca2 if effects of Ca2 on

3H-ouabain binding are to be examined. Cultured cells have the disadvantage of

originating from embryonic tissue which may have very different properties from
adult tissue. Additionally, cell cultures usually exist as a mixed population of
muscle and non-muscle cell types and while in culture the properties of the cells,
including the number of receptors for drugs, may change.

2 3

In order to examine effects of Ca + on H-ouabain binding and sodium-pump

2+-tolerant myo-

activity, improved techniques for isolation and selection of Ca
cytes have been developed. The preparations we use contain greater than 8096
rod-shaped cells. The remainder of the cells are dead or dying as determined by a
rounded appearance. Although these preparations are not homogeneous, they are
currently the best available. Direct effects of interventions on cardiac glycoside

binding and sodium-pump activity should be discernable using this myocyte

preparation.

E. Objectives

The objective of this study was to develop a preparation with which direct
effects of agents or interventions on cardiac glycoside binding to Na,K-ATPase

and on sodium—pump activity could be examined in intact cardiac muscle cells.

This preparation was then to be used to examine effects of intracellular Ca2+ and

3

sarcolemmal phosphorylation on H-ouabain binding and sodium-pump activity.

MATERIALS AND METHODS

A. Myocyte Isolation

 

Isolation of viable myocytes from mammalian heart in high yield is a
difficult task which on several occasions during the course of these experiments
seemed to be impossible. Reasons for variable degrees of success could often be
attributed to the collagenase used for the digestion. The protocol for isolation as
written here has evolved over two years of experiments, during which time
numerous variations were attempted to increase yield and viability of myocytes.
The basic procedure, however, has remained intact over the entire time. This
procedure for myocyte isolation has been used successfully to isolate viable
myocy tes in high yield from hearts of guinea-pig, rat, ferret and rabbit.

Myocytes were isolated from guinea-pig hearts by digestion with collagenase
plus hyaluronidase. Male animals weighing 450 to 600 g were injected with
heparin (700 units, i.p.) and sacrificed 40 min later by a sharp blow to the neck.
Heart and lungs were quickly removed and immediately perfused with a modified
Krebs-Henseleit bicarbonate (KHB) buffer solution at 37°C with a constant flow
rate of 6.5 ml/min. The composition of KHB solution was 118 mM NaCl, 27.1 mM

NaHCO3, 2.8 mM KCl, 1 mM KH2P04, 1.2 mM MgSO4, 1.8 mM CaCl 2.5 mM

2,
Na-pyruvate and 10.0 mM dextrose. The solution was saturated with a 9596 Oz,
596 002 gas mixture yielding a final pH value of 7.4. We found it important to
prepare the heart so that circulation would be maintained during perfusion with

2+

solutions containing low concentrations of Ca . This was achieved by leaving as

much of the free aorta as possible below the inlet cannula, cutting the pulmonary

12

13

artery close to its origin and occluding the pulmonary veins with a silk suture tied
between the heart and lungs before removal of the lungs. After a 15 min
perfusion with KHB buffer, perfusion solution was changed to one containing 105.1
mM NaCl, 20.0 mM NaHCO3, 2.8 mM KCl, 1 mM KH2P04, 1.2 mM MgSO4, 0.01
mM CaC12, 5.0 mM mannitol, 10.0 mM taurine (2-aminoethanesulfonic acid), 10.0
mM dextrose, 5.0 mM Na—pyruvate and saturated with a 95% 02, 5% CO2 gas

2+ solution, contractions ceased

2+

mixture. During the perfusion with this low Ca
and left atria became swollen. Time of perfusion with low Ca solution before
addition of enzymes was usually 8 min but was occasionally decreased to as little
as 5 min to enhance subsequent digestion.

Following the initial perfusion with a low Ca2+ solution, perfusion with low

Ca2

+ solution containing collagenase (0.52 mg/ml) and hyaluronidase (0.2 mg/ml)
was started. Effluent from the hearts was discarded during the first 3 min of
perfusion with solution containing enzymes. After the first 3 min, the effluent
was collected and recirculated. During this perfusion, hearts were immersed in
the recirculating medium and ﬂow-rate was reduced to 5 m1/min for each heart.
Perfusing solution was continuously bubbled with a 95% 02, 5% CO2 gas mixture
during the 52 min digestion period.

Subsequent to digestion, ventricular muscle was excised from each heart and
minced into 12 ml of a solution composed of 3 parts KHB solution and 1 part fresh

low Ca2+

solution containing collagenase and hyaluronidase and maintained at
37°C. The minced muscle (small chunks of 1 to 2 mm diameter and a fluffy layer
of single cells and cell aggregates) was permitted to sediment and most of the
supernatant solution was aspirated away. Additional KHB solution (4 ml, 37°C)

was added to the sediment of minced muscle. Fresh KHB solution (4 ml, 37°C)

was added twice more without removal of supernatant. Between additions of KHB

14

solution, sedimented materials were drawn in and out of a wide mouthed pipet to
disaggregate the myocy tes.

Following disaggregation, cell suspensions from individual hearts were
filtered through a stainless steel mesh (pore Size, 400 um) and resulting
suspensions were diluted to a final volume of approximately 30 ml with KHB
solution containing 1.8 mM CaClz. The suspension was centrifuged at 60 x g for
15 sec. Supernatant solution was discarded and rod-shaped myocytes were
selected for by gravity sedimentation through KHB solution. Dispersed cells were
separated from cell aggregates and tissue debris by filtration through a nylon
mesh (pore size, 200 um). Myocytes passing thrugh the nylon mesh were collected
by gravity sedimentation and loaded into a glass column (29 cm long; 0.8 cm
internal diameter). The column was adjusted to an angle of 2 to 5 degrees from
vertical. Flow of KHB solution at a rate of 3.0 ml/min from the bottom to the
top of the column maintained viable cells within the column. Viable cells
preferentially aggregated into loose clusters which filled 40 to 60% of the column.

Cells were maintained in the column for 60 min, during which time the flow
rate was periodically increased to 10 ml/min for Short periods to prevent the cells
from forming a packed pellet at the bottom of the column. Myocytes were
removed from the column 10 min before the start of an experiment, and rod-
shaped cells were selected for by 2 to 3 gravity sedimentations in KHB solution.

The single most important part of myocyte isolation is to ensure that tissue
digestion is sufficient so that minimum mechanical disaggregation is necessary.
Best results were obtained when hearts were digested to such an extent that some
hearts would fall from the cannula when the support of the enzyme solution they
were immersed in was removed.

Selection of collagenase was by trial and error. Samples from all available

sources were tested using a single protocol. The collagenase producing the best

15

results was purchased and the digestion procedure "fine tuned" for optimal results
with that batch of enzyme. The following points were modified to enhance
digestion: time of perfusion with low Ca2+ solution prior to addition of enzymes;
time of recirculating perfusion with enzymes; and volume of recirculating
perfusate with enzymes. Increasing the time of recirculating perfusion with
enzymes always increased digestion of the heart, but this was avoided because

extended time in low Ca2+

2

solution appeared to decrease cell viability. Time of
perfusion with low Ca + solution prior to addition of enzymes was decreased from
the standard 8 min to as little as 5 min to enhance subsequent digestion. The
mechanism responsible for enhancement of digestion is unknown but may be a
decrease in time for efflux of some factor from the heart before the effluent is

collected for recirculation. This same mechanism may also be involved in

enhancement of digestion when volume of recirculating perfusate is reduced.

B. 3H—Ouabain Equilibrium Bindﬂsmdies

 

Specific binding of 3H—ouabain to Na,K-ATPase from cardiac muscle has
been shown to correlate well with inhibition of the enzyme (Erdmann e_t a_l_., 1976).
Specific binding of 3H-ouabain causes inhibition of the sodium-pump in intact
tissue and isolated myocyte preparations (Werden e_t a_1., 1983). Inhibition of the
sodium-pump by cardiac glycoside binding to the Na,K-ATPase is believed to be
responsible for the positive inotropic effect of these agents (Akera e_t a_1., 1974).
Binding of 3H—ouabain to cardiac myocytes was examined to characterize the
binding of this glycoside to the Na,K-ATPase in intact cardiac muscle cells and to

determine if affinity of the Na,K-ATPase or number of 3

H—ouabain binding sites
in intact cells could be altered by acute or chronic interventions.
Affinity of binding sites for 3H—ouabain and the number of binding sites

were estimated using the method of Akera and Cheng (1977) or by analysis of

l6

Scatchard plots. For analysis by the method of Akera and Cheng, myocytes (0.2-
0.6 mg protein) were incubated in 1.2 to 2.0 ml of KHB solution or homogenates
(0.3-0.4 mg protein), were incubated in 1.5 ml of a solution containing 1.0 mM
MgClz, 1.0 mM Tris-phosphate, and 10.0 mM Tris-HCI (pH 7.5 with Tris-base).
Homogenates were prepared from myocytes which had been frozen. Myocytes
were homogenized by several strokes with a Dounce ball-type homogenizer.
Incubation tubes were prewarmed to 37°C and maintained at that tempera-
ture for the duration of the incubation (35-90 min). Solutions for cell incubation
were saturated with 95% 02’ 5% CO2 before addition to incubation tubes. Tubes
containing cells were filled with the same gas mixture and sealed with marbles.
Solu tiors for homogenate incubation were not exposed to any gas except room air.

Each incubation tube contained 3

H-ouabain (50 or 100 nM, final concentra-
tion), and various concentrations of non-labeled ouabain (0, 0.05, 0.1, 0.3, 1.0, 3.0
or 300 11M) were present in equal numbers of tubes. After incubation, the binding
reaction was stopped and cells were collected on GF/C or homogenates were
collected on GF/B fiberglass filters (Whatman). Specific 3H-ouabain binding was
calculated by subtracting the binding observed in the presence of 300 uM non-
labeled ouabain. Affinity (KD) and 3H-ouabain binding site number (Bmax) were
then estimated from linear plots on logarithmic probability paper describing the
inhibition of specific 3H-ouabain binding by the non-labeled ouabain.

For determination of kinetic parameters of 3H—ouabain binding by Scatchard
analysis, myocytes (0.2-0.4 mg protein) were incubated in 1.2 ml of KHB solution
containing various concentrations (0.05, 0.1, 0.2, 0.5, 1.0, 2.0, 3.0, 5.0, 10.0 or

3H-ouabain for

20.0 11M) of 3H-ouabain. Cells were incubated in the presence of
60 min then either collected by a centrifugation method, which is described in
detail later, or resuspended in a 100-fold volume (30 ml) of KHB solution

containing a concentration of non-labeled ouabain equal to the concentration of

17

3H—ouabain the cells had been incubated with and 13.8 mM KCl. Dissociation of

3H-ouabain was allowed to proceed for 60 min then cells were collected by
centrifugation. Samples of cells collected immediately following the 60 min

incubation with 3

H-ouabain or after that incubation followed by a 60 min
incubation for dissociation of bound drug were analyzed for determination of total
binding and non-specific binding, respectively. Specific binding to the Na,K-

ATPase was calculated as the difference in these values.

C. Dissociation of 3H-Ouabain from Myocytes

 

3

Differential rates for dissociation of bound H—ouabain from Specific and

non-specific binding sites allow for estimates of the relative incorporation of 3H-

ouabain into each of these sites in whole tissue by monitoring dissociation after

3

binding (Kjeldsen gt g1., 1985). Relative amounts of H-ouabain bound to

different conformations of Na,K-ATPase can also be determined by monitoring

dissociation of drug bound to the isolated enzyme (Choi and Akera, 1976).

3

Dissociation of H-ouabain from myocytes was examined to determine the

3H—ouabain can

relative amount of non-specific binding and to substantiate that
bind to more than one conformation of the Na,K-ATPase in intact myocytes.

Time courses for dissociation of 3H-ouabain from myocytes were monitored
following incubation of cells in 0.2, 0.5 or 2.5 uM 3H—ouabain for 40 or 60 min.
Cell pellets were collected by gravity sedimentation after the binding reaction

3H-

and sampled by centrifugation at 1500 xg for 2 min to determine total
ouabain bound before the initiation of dissociation. The remainder of the cell
pellet was dispersed in a 100- or 200-fold volume of KHB solution containing a

3H—ouabain

concentration of non-labeled ouabain equal to the concentration of
the cells had been incubated with. Concentration of K+ in the binding and

dissociation media were varied as described for individual experiments. After

18

resuspension of myocytes to initiate dissociation 6.0 ml of well-mixed cell
suspension was removed at regular time intervals for up to 3 hours. Aliquots of
cell suspension were taken with an adjustable volume pipet and cells collected by
filtering the suspension through Silanized GF/C fiberglass filters (Boehringer

Mannheim). Supernatants from the KHB solution in which cells had been

3

incubated for binding of H-ouabain were centrifuged for 2 min at 1500 x g to

remove cells. These supernatants were then added to an aliquot of solution
identical to the dissociation medium in proportion to the volume of cells which
had been added. This solution was then sampled four times in a manner identical

to that used for sampling cell suspensions. The mean amount of radioactive

material retained on the filters was used as an estimate of the background 3H-

ouabain binding to filters. Dissociation rates from different sites were deter-

mined to be linear exponential processes by use of a curve-peeling procedure or by

3

subtracting the H-ouabain remaining bound after 60 min of dissociation from the

amount remaining bound at all earlier time points. Dissociation rate constants

(k-l) were estimated from the slopes of the linear exponential processes.

86

D. Rb+ Up take Studies

Sodium-pump activity was estimated in myocytes as the difference in 86Rb+
uptake observed in the absence and presence of 1 mM ouabain. Myocytes used in

86Rb+ uptake studies were prepared in the normal manner except that a modified

KHB solution containing 5.0 mM RbCl and 1.0 mM NaH2P04 was used instead of

one containing 2.8 mM KCl and 1.0 mM KH2P04 from the time cells were loaded

86

on columns for the 60 min elutriation. Uptake of Rb+ was initiated by adding

cells to prewarmed (37°C) incubation tubes containing 2.0 ml of the modified KHB

6 86

solution with 5.0 mM RbCl and tracer amounts of 8 Rb+. Non-Specific Rb+

uptake was estimated as the uptake of 86Rb+ not inhibited by the presence of 1

19

mM ouabain. Exposure of myocytes to ouabain and 86Rb+ were simultaneous

when estimates of non-specific uptake were made. After 6 min of exposure to

86Rb+, myocytes were collected by centrifugation and then sampled for determi-

86 f 86

nation of Rb+ and protein content. Specific activity 0 Rb+ was measured

directly by sampling the incubation media after removing cells.

Sodium—pump capacity can only be measured when the intracellular Na+
concentration is not the factor limiting sodium-pump activity (Akera gt a_1., 1981).
Capacity of the sodium-pump was examined using cells which had been preincu-
bated for 5 min with maximally effective concentrations of the Na+-ionophore
monensin, or using cells which had been sodium-loaded by incubation in medium

devoid of K+ and Rb+. These interventions each increase sodium-pump activity as

86

estimated by ouabain sensitive Rb+ uptake.

86

Ouabain sensitive Rb+ uptake was estimated using KHB solutions contain-

ing 2.0 mM RbCl when determining sodium-pump capacity. For these experiments
the cells were preincubated with monensin. Myocytes were added to prewarmed

(37°C) incubation tubes containing monensin. After a 5-min incubation in the

86Rb+ were added to the incubation

tubes. Cells were collected by centrifugation after a 3-min exposure to 86Rb+

86

presence of monensin, tracer amounts of

then sampled for determination of Rb+ and protein content. Non-specific

86Rb+ uptake was estimated by adding 1 mM ouabain (final concentration) in

solution with the 86Rb+.

Myocytes were sodium-loaded by elutriation in a glass column with KHB
solution devoid of K+ and Rb+ and containing only 10 11M Ca2+. Cells were
maintained in the column at 37°C with a continuous flow (3 ml/min) of fresh

solution to remove any K+ or Rb+ released by the cells. Under these conditions,

Na+ accumulates in the cells because a counter-ion to support efflux of Na+ from

20

myocytes by the Na,K-ATPase is not available. Effectiveness of the sodium-

loading procedure was determined by removing cells from the column at 15 min

86

intervals and measuring ouabain-sensitive Rb+ uptake as described earlier, with

incubations taking place in 5.0 mM Rb+ KHB solution containing tracer amounts

of 86 86

Rb+. Incubations were 2 min in duration for Rb+ uptake into Na+-loaded

cells.

Reserve capacity of the sodium-pump is decreased by interventions which
increase the Na+ concentration inside cells. Concentration dependence for
ouabain inhibition of sodium-pump activity in non-stimulated myocytes or myo-
cytes exposed to 50 11M monensin was determined to demonstrate that increased
Na+ influx reduces sodium-pump reserve capacity. Myocytes were exposed to 0.1,

0.3, 1.0 or 3.0 uM ouabain for 60 min before initiation of 86Rb+ uptake in

86

O O O O O O O O +
solutions contaInIng the same concentration of ouabaIn. Ouabain-senSItIve Rb

uptake was measured as described earlier with incubations taking place in 5.0 mM

86

Rb+ KHB solution with tracer amounts of Rb+ for 6 min.

E. S_epara tion of Myocy tes and Bound Ligands from Non-bound Ligands
Rapid filtration and centrifugation methods were utilized to collect myo-

cytes and separate bound from non-bound ligands. Collecting samples by rapid

3

filtration is the standard method used in studies examining H-ouabain binding to

isolated enzyme or tissue homogenate preparations. This method also worked well

for collection of isolated myocy tes after 3

3

H-ouabain binding. Myocy tes incubated
with H-ouabain were poured onto fiberglass filters after stopping the binding
reaction by the addition of 5 m1 of an ice-cold solution, comprised of 0.1 mM non-
labeled ouabain, 15 mM KCl and 50 mM Tris-HCl buffer (pH 7.5). Filters and

myocytes were washed twice with 5 m1 volumes of the same solution used to stop

the reaction. Cells were always collected on fiberglass filters. Effective pore

21

size and preconditioning of filters with dimethyldichlorosilane varied according to
the experiment. Myocytes trapped on filters were digested with 0.6 ml of tissue
solubilizer (Protosol, NEN). Scintillation fluid (9 ml) using PPO (2.5 diphenyloxa—
zole) and POPOP (1,4-bis [2—(4—methyl-5-phenyloxazolyl)]-benzene) as scintillants
was added to each vial and radioactivity was assayed using a liquid scintillation
spectrophotometer. Counting efficiency (approximately 30%) was monitored by
the external standard channel ratio.

Collection of myocytes by centrifugation was a necessity for experiments

86Rb+ uptake because background radioactive material trapped by

86

examining
fiberglass filters in these experiments often exceeded the total Rb+ in samples
of myocytes. The centrifugation method was also advantageous because both
label and protein were determined after the binding or uptake reaction was
complete. Therefore, protein concentration in incubation tubes could vary
somewhat. Protein concentration in incubation tubes was difficult to control
because myocytes settle quickly in KHB solution and repeated agitation in a
manner vigorous enough to uniformly disperse cells resulted in damage to
myocytes.

Centrifuge tubes for sampling were prepared by pipeting 4.0 ml of a sucrose
solution containing 280 mM sucrose, 10.0 mM procaine HCl and 0.9 mM CaCl2
into 15 ml polypropylene tubes (Bec ton-Dickinson), then slowly adding 9.0 ml of a
less dense solution containing 99.0 mM NaCl, 30.0 mM RbCl, 20.0 mM procaine
HCI, 10.0 mM HEPES (N—2—hydroxyethylpiperazine—N'-2—ethanesulfonic acid), 2.0
mM BaClz, 0.9 mM CaClZ, 0.1 mM ouabain and a trace of Patent Blue Violet dye
(pH 7.1 with 2 M NaOH). The less dense solution was added slowly so at least 3 ml
of the sucrose layer remained unmixed with the less dense layer. These "sampling
tubes" were placed in a NaCl-ice bath with the temperature adjusted to -2°C and

were allowed to stand at least 1 hour before use.

22

86

Prior to sampling, myocytes incubated with 3H-ouabain or Rb+ were

allowed to settle in incubation tubes to as great an extent as possible within the
confines of individual experiments. At the time 3H-ouabain binding or 86Rb+
uptake reac ﬁons were to be stopped, 0.5 ml of soluﬁon containing cells was taken
from the bottom of incubation tubes and added to a centrifuge tube prepared as
described earlier. Homogeneous dispersion of the sample is calculated to increase
the temperature of the less dense soluﬁon to 0°C. Therefore, all reac ﬁons
involving Na,K-ATPase would be virtually stopped. Preliminary experiments

3H-ouabain or 86

showed Rb+ content of myocytes did not change significantly
when centrifugaﬁon was delayed for up to 15 min after samples were added to
centrifuge tubes.

Cells were collected by centrifuging the samples for 2 min at 1500 x g in a
refrigerated centrifuge maintained at 2°C. The less dense soluﬁon was identified
by its blue color. The blue solu ﬁon containing non-bound 3H-ouabain or 86Rb+
was aspirated from the centrifuge tube leaving approximately 3 ml of sucrose
solution covering the cell pellet. Tubes with cell pellets were frozen rapidly by
immersing them in methanol at -30°C. Cell pellets were collected by cutﬁng the
bottom from the frozen tubes and then analyzed to determine content of
radioac ﬁve material and protein.

f 86Rb+ content were analyzed without further

3

Samples for determinaﬁon 0
treatment using a gamma counter. Samples for determinaﬁon of H-ouabain
content were dried overnight at room temperature then wetted with 75 ul of
disﬁlled water before digesﬁng and counﬁng as described for filtered samples.

Wetﬁng of samples was necessary to maintain the sucrose in soluﬁon after

addiﬁon of scinﬁllaﬁon fluid.

23

F. Silane Treatment for Fiberglass Filters
3

 

Esﬁmates of non-specific binding of H-ouabain to myocytes and fiberglass
filters were large and varied to an extent that it was difficult to have confidence
in these esﬁmates. Accurate esﬁmaﬁon of non-specific 3H-ouabain binding to
filters was parﬁcularly important when determining non-Specific 3H—ouabain
binding to cells. Silane treatment of filters reduced average non-specific binding
to GF/C filters from Boehringer Mannheim only slightly but significantly reduced
variaﬁon in 3H-ouabain binding to the filters.

Silanizaﬁon was accomplished in the following manner. Approximately 100
filters were stacked on a stainless-steel mesh inside a 50-ml plasﬁc syringe fitted
with a stopcock. The syringe was suspended verﬁcally and a second mesh placed
on top of the filters. Dimethyldichlorosilane (1% in hexane) was poured over the
filters unﬁl filters and mesh overlay were immersed. The Silane soluﬁon was
pressed out of filters 30 min after its addiﬁon and the residual Silane was removed
by two washes with hexane. Following each hexane wash, the hexane was also
pressed out of the filters. Methanol was poured over the filters after the second
rinse with hexane. Filters remained immersed in methanol for 30 min then
methanol was pressed out of the filters and a final hexane rinse was done in a
manner idenﬁcal to the iniﬁal hexane rinse. Filters were dried under vacuum and
those in contact with the stainless steel mesh during the silanizaﬁon procedure

were discarded.

This procedure could not be used for Whatman fiberglass filters.

G. Experiments with K+-Deficient Animals

 

Changes in the density of ac ﬁve sodium-pump sites or acﬁvity of the Na,K-
ATPase may occur in animals with reduced serum K+ concentraﬁon (Ward and

Cameron, 1984). To substanﬁate that the sodium-pump was operaﬁng at capacity

24

in myocy tes exposed to high concentraﬁons of monensin or following loading with
Na+ in soluﬁons devoid of K+ and Rb+, attempts were made to increase sodium-
pump capacity by increasing the density of Na,K-ATPase by depleﬁng animals of
K+.

Guinea pigs were depleted of K+ by maintaining the animals on a K+-
deficient diet. Animals were housed in individual cages and were fed a diet low in
K+. Control animals were fed a diet of the same composiﬁon supplemented with
K+. Animals were maintained on either of these diets for 2 to 3 weeks before
hearts were removed for cell preparaﬁon.

Depleﬁon of K+ was assessed by measuring acﬁviﬁes of Na+ and K+ in
serum using a Na+ and K+ acﬁvity analyzer (Orion). Serum was prepared from
blood drawn from ketamine (40 mg/kg, i.m.) anestheﬁzed animals.

Kineﬁc parameters for 3

H-ouabain binding to myocytes were determined by
the method of Akera and Cheng (1977) as previously described. Sodium-pump
acﬁvity in the absence and presence of monensin was measured as previously

described.

H. Miscellaneous Methods

 

Protein determinaﬁons were by the method of Bradford (1975) using a dye
concentrate purchased from Bio-Rad. This method was necessary due to the
variable sucrose content in samples obtained by the centrifugaﬁon method.
Sucrose has been shown to interfere with other procedures for the determinaﬁon
of protein such as the Lowry and the Biuret procedures (Bradford, 1975).

Staﬁsﬁcal analysis was by grouped t—test, paired t-test or analysis of

variance. Significance was established by a P value of < 0.05.

25

1. Materials
Triﬁum-labeled ouabain (generally labeled, specific radioac ﬁvity 20.0

86RbCl (concentraﬁon and specific radioacﬁvity variable) were

Ci/mmol) and
purchased from New England Nuclear, Boston, MA.

Collagenase was purchased from Cooper Biomedical, Malvern, PA, Boeh-
ringer Mannheim, Indianapolis, IN, and Sigma Chemical Company, St. Louis, MO.
Hyaluronidase, dime thyldichlorosilane and Pa tent Blue Violet dye were all pur-
chased from Sigma Chemical Company.

Filters were purchased from Boehringer Mannheim (fiberglass filters, 24
mm), or Fisher Scienﬁfic Company (Whatman GF/C and GF/B fiberglass filters),
Pittsburgh, PA.

Tissue solubilizer was purchased from New England Nuclear. Dye reagent
concentrate for protein analysis was purchased from Bio-Rad Laboratories,
Richmond, CA. Polypropylene centrifuge tubes (Bec ton-Dickinson) were pur-
chased from Sargent-Welch, Livonia, MI. Heparin sodium (from beef lung) was

purchased from the Upjohn Company, Kalamazoo, MI. The K+-deficient diet was

purchased from BioServ Inc., French Town, NJ.

RESULTS

A. Isolaﬁon of Myocytes from Guinea-pigHeart
2+

 

Isolaﬁon of Ca tolerant myocytes from mammalian heart has been
described by many invesﬁgators (eg: Farmer gt gl_., 1977; Isenberg and Klockner,
1982). The procedure described here which was used to isolate myocytes from
guinea-pig heart is not substantially different from others which have been
published. Populaﬁons of myocytes used for experiments contained greater than
80% of cells which were rod-Shaped when incubated in soluﬁons containing
millimolar concentraﬁons of Ca2+. Rounded cells in the preparaﬁons accounted
for less than 20% of all cells and were apparently incapable of maintaining a low

intracellular Ca2+

concentraﬁon, as shown by their hypercontracted state.
Approximately 50% of rounded cells excluded the dye Trypan Blue; therefore, it is
possible that some rounded cells had not completely lost membrane integrity.
Photographs of typical myocytes are shown in Figure 1. Rod-shaped cells have
clear striaﬁons exclude trypan blue, have normal resﬁng membrane potenﬁal2 and
respond to electrical Sﬁmulaﬁon with discrete contracﬁons. These features
indicate that the cells have intact membranes, are alive and have normal
responses to sﬁmuli. Details on the surface of these myocyes were examined by

transmission electron microscopy.E An electron photomicrograph of one cell is

shown in Figure 2.

 

EMembrane potenﬁals of myocytes were measured using microelectrodes by
Robert W. Hadley working in the laboratory of Dr. Joseph R. Hume at Michigan
State University.

26

27

 

Figure 1. Photograph of ventricular myocytes isolated from guinea-pig heart by
treatment with collagenase and hyaluronidase. Typical preparaﬁons had gggater
than 80% of cells rod—shaped and quiescent in solu ﬁon containing 1.8 mM Ca .

28

 

Figure 2. Electron micrograph of a single cardiac myocyte. Myocytes were fixed
in 10% glutaraldehyde solu ﬁon then prepared for examinaﬁon by seaming
electron microscopy using standard techniques.

29

Myocytes were isolated and rod-Shaped cells selected as described in
Methods. Purificaﬁon of myocytes was repeated unﬁl microscopic examinaﬁon
indicated greater than 80% of the cells in the preparation were quiescent and
maintained a rod shape. This usually required two or three sedimentaﬁons of
myocytes through KHB soluﬁon at unit gravity. Further selecﬁon of rod-shaped
cells by gravity sedimentaﬁon is possible but was usually not attempted due to
excessive loss of ﬁssue. Isolated myocytes were very sensiﬁve to mechanical
sﬁmuli. The fragile nature of these cells precludes more vigorous methods for
selecﬁon of rod—shaped cells. The relative percentage of rod-shaped cells in
preparaﬁons was very consistent due to it being controlled. Also, each populaﬁon
of myocytes acted as its own control as experiments were designed so each
intervenﬁon or concentraﬁon of drug was tested in every populaﬁon of myocytes
used for a particular experiment. Therefore, no attempt was made to quantitate
the percentage of rod-shaped cells in individual preparaﬁons unless the viability
of cells was being examined.

1. Viability of Guinea-Pig Myocytes Incubated in the Presence or Ab-
sence oﬁuabain

Equilibrium of ouabain binding to Na,K-ATPase from guinea-pig heart
requires greater than a 30-min incubaﬁon of the glycoside with the enzyme as
judged by the inotropic response of guinea-pig atrial muscle to ouabain (Ku gt 9.1.,
1976). It is well documented however, that high concentraﬁons of cardiac
glycosides are toxic to cardiac muscle preparations. To determine if myocytes
isolated from guinea—pig heart were stable in soluﬁons containing ouabain, cells

were incubated for 60 min in KHB soluﬁon containing 1.0 uM or 1.0 mM ouabain.

 

l3Elec tron microscopy was performed by Vivion E. Shull and Dr. Karen Baker
at the Electron Opﬁcs Center at MSU.

30

Myocytes from the same preparaﬁon were incubated for 60 min in KHB soluﬁon
without ouabain or sampled immediately after addiﬁon to an incubaﬁon soluﬁon.
Myocytes were sampled by dispersing 0.3 m1 of cell suspension into 2.0 ml of 10%
formalin. Dispersing of myocy tes in formalin caused cells to be fixed in either a
rounded or a rod shape. Fixed cells were counted using a hemocytometer.
Relaﬁve amounts of rod-Shaped and rounded cells were determined after counﬁng
at least 100 cells.

Incubaﬁon of guinea-pig myocytes in KHB soluﬁon at 37°C for 60 min
caused a significant decrease in the percentage of cells retaining a rod shape
(Figure 3). Myocytes incubated with 1.0 uM ouabain, however, showed no loss of
viable cells as determined by maintenance of the rod shape. Since a concentra-
ﬁon of 1.0 uM ouabain can be toxic to preparaﬁons of contrac ﬁng cardiac muscle
from guinea pig, the protec ﬁve effect on myocy tes is surprising. It is conceivable
that this reﬂects a decreased energy demand on the cells due to inhibiﬁon of
some pump units, or a decrease in Na+ inﬂux because the cells are quiescent.
Complete inhibiﬁon of the sodium-pump by 1 mM ouabain caused almost complete
loss of viability after a 60 min exposure (Figure 3). Microscopic examinaﬁon of
these cells prior to fixing in formalin Showed that cells sﬁll retaining a rod shape
were contracﬁng spontaneously. The contracﬁons were not discrete but more

closely resembled a writhing moﬁon indicaﬁng that the cells were dying.

B. Bindingof 3IJI-Ouabain to Myocytes Isolated from GuineajiLHeart
lnhibiﬁon of Na,K-ATPase is believed to be responsible for the posiﬁve

inotropic effect of cardiac glycosides on cardiac muscle (Akera e_t a_1., 1970; Ku e_t

a_1., 1976). Binding of cardiac glycosides to the Na,K-ATPase results in inhibiﬁon

of this enzyme (Matsui and Schwartz, 1968; Gelbart and Goldman, 1977). Kineﬁc

31

Stability of Myocyte Preparations
90 - ‘

80 '
7O —
60 r

4
*

 

 

T
_L

 

50 ”
40 r
30 '
20 '
10 .

Rod-shaped Cells 2 of Total

96!!
z IImmn___
Initial 60-min 1 MM 1 mM

KH Ouabain Ouabain

Incubation Condition

 

 

 

 

W‘r

 

 

 

 

Figure 3. Viability of myocy tes from guinea-pig heart in the presence of ouabain.
Myocytes were isolated and rod-shaped cells were selected as described in the
text. Bars represent the percentage of myocytes in the enﬁre populaﬁon which
were rod-Shaped at iniﬁaﬁon of incubaﬁon El , following a 60—min incubaﬁon

Q , following a 60—min incubaﬁon with 1.0 um ouabain , following a 60-
min incubaﬁon with 1 mM ouabain m1] . All incubaﬁons were in KHB soluﬁon
containing 1.8 mM Ca . Percentages were calculated after counﬁng at least 100
cells which had been fixed by being sampled into 10% formalin at the indicated
ﬁme (n=5). at: significantly different from the percentage rod-Shaped at the
iniﬁaﬁon of incubaﬁon, {significantly different from the percentage rod-Shaped
after a 60-min incubaﬁon in KHB soluﬁon. Significance (p< .05) determined by
analysis of variance with individual comparisons by method of least Significant
differences.

32

parameters of binding (associaﬁon rate constant, k1

; and dissociaﬁon rate con-
stant, k'l) determine the extent of enzyme inhibiﬁon at any given concentraﬁon
of a cardiac glycoside and may be subject to modificaﬁon by many factors. These
factors would, therefore, contribute to the regulaﬁon of the pharmacological
effects of the glycoside.

Many factors which affect glycoside binding to Na,K-ATPase can be studied
using preparaﬁons of isolated enzyme. Other factors may be lost or inacﬁvated
during preparatory procedures or ﬁssue homogenizaﬁon and therefore, must be
examined in intact cells. Possible binding sites for glycosides other than the
Na,K-ATPase may also be lost or exposed during ﬁssue disrupﬁon and the ques ﬁon
of their existence must be addressed using intact cells. Another advantage of
intact cells over ﬁssue fracﬁons for examining glycoside binding to Na,K-ATPase
is the maintenance of ionic gradients across the sarcolemmal membrane. This is
advantageous because ionic environments on the internal and the external sides of
the sarcolemma can be changed individually. Also, living myocytes contain an
ATP regeneraﬁng system which ensures an adequate supply of substrate for the
ATPase. The presence of ionic gradients in intact cells is an aspect of this
preparaﬁon which must be considered during the design of experiments. In
comparison to isolated enzyme preparaﬁons, the presence of ionic gradients and
possible changes in ionic gradients subsequent to sodium-pump inhibiﬁon in intact
cells, can confound interpretaﬁon of binding data.

Examining glycoside binding in isolated myocytes is a vastly superior
technique to examining the binding in intact cardiac muscle. This is evident from

a comparison of published reports of 3

H—ouabain binding to intact cardiac muscle
with data reported here. Advantages of myocytes over whole ﬁssue are the

presence of only one cell type, the lack of diffusion barriers and a much reduced

33

amount of non-specific binding. Esﬁmaﬁon of kineﬁc parameters for 3H—ouabain

2+ tolerant myocytes isolated from guinea-pig heart was

binding to receptors in Ca
done to characterize the binding of this cardiac glycoside to its receptor(s) in
intact cells. Possible existence of regulatory mechanisms for 3H-ouabain binding
unique to intact ﬁssue was then studied by incubating cells with agents known to
affect the contracﬁle state of cardiac muscle.

1. Esﬁmaﬁon of Non-Specific I—l-Ouabain Bindingto GuineaﬂMyocytes

Accurate esﬁmaﬁon of kineﬁc parameters for 3H—ouabain binding‘to
Na,K-ATPase requires adequate ligand concentraﬁons and accurate esﬁmation of
non-Specific binding. Cofactors for the Na,K—ATPase, besides the substrates Na+
and K+, are ATP and Mg2+. These ligands are normally maintained at adequate
concentraﬁons in viable cells. For Scatchard analysis of binding data, myocytes
must be incubated in a range of 3H—ouabain concentraﬁons so that occupancy of 5
to 95% of the binding sites will be described (Burgisser, 1984). Because the
sodium—pump is inhibited by ouabain binding to Na,K-ATPase and total inhibiﬁon
of the sodium-pump destroys the myocytes, equilibrium of the binding reacﬁon
with 95% occupancy of binding sites is probably not feasible.

A potenﬁally more serious problem is the necessity for accurate
esﬁmates of non-specific binding. Because an effecﬁve antagonist for glycoside
binding which does not itself inhibit the Na,K-ATPase is not available, an excess
of non-labeled drug is included in reacﬁon mixtures when determinations of non-
specific binding to isolated enzyme or ﬁssue homogenate preparaﬁons are made.
Myocytes incubated with concentraﬁons of ouabain sufficient to block specific
binding of labeled drug undergo an extreme morphological change within 60 min
(Figure 4). Media in which these cells are incubated become yellow, indicaﬁng
that some cell lysis has occurred. The morphological changes resulﬁng from

3

ouabain toxicity can potenﬁally affect non-specific binding of H-ouabain,

34

 

0-min
Toxic

ig heart following a 6
and 1.0 mM ouabain.
t this concentraﬁon cause morphological changes in the cells.

1.8 mM Cay?

mg

Photograph of myocytes from guine
KHB soluﬁon contain

ubaﬁon in

Figure 4.
inc

effects of ouaba

Ina

mg the cells.

ith microscopic examinaﬁon and by a

aﬁon medium contain

Incub

Lysis of some myocytes was evident w

yellow color of the

35

causing it to be under- or overesﬁmated. Another consideraﬁon when esﬁmaﬁng

3 3H-

non-specific binding of H—ouabain to intact cells is the potenﬁal uptake of
ouabain into the myocytes as a component of non-specific binding. The Na,K-
ATPase has been proposed to transport cardiac glycosides into cardiac cells
(Dutta e_t g_l., 1968; Fricke and Klaus, 1977). Any carrier-mediated inﬂux of 3H-
ouabain would be blocked if excess non-labeled ouabain was used when making
determinaﬁons of non-specific 3H—ouabain binding. Therefore, specific binding
would be overesﬁmated. For these reasons it was necessary to substanﬁate that
the amount of 3H—ouabain retained in samples of myocytes and not specifically
bound was esﬁmated correctly.

Kineﬁcs of the dissociaﬁon of 3H—ouabain from myocytes were
examined in an attempt to estimate non-specific binding. Myocytes were
incubated in a non-toxic concentraﬁon of 3H-ouabain (0.2 11M) for 60 min.

Dissociaﬁon of bound 3

H-ouabain was iniﬁated after collecﬁng the myocytes by
gravity sedimentaﬁon. Myocytes were resuspended in a 100-fold volume of KHB
soluﬁon containing an idenﬁcal concentraﬁon of non—labeled ouabain. This
suspension was sampled at various time points and bound ligand separated from

3

non-bound ligand by rapid filtraﬁon. Dissociaﬁon of bound H-ouabain from

3H-Ouabain associated

guinea-pig myocytes occurred in two phases (Figure 5).
with the site from which dissociaﬁon was rapid accounﬁng for 84.6:1.1 percent
(n=6) of bound drug.

A plot of these data on a semilogarithmic scale produced an upwardly
concave curve describing the first sixty minutes of dissociaﬁon (Figure 6). The
porﬁon of the curve describing dissociaﬁon at ﬁme points greater than 60 min
appeared to become a straight line parallel to the X-axis. The esﬁmates of 3H-

ouabain bound to myocytes after greater than 60 min of dissociaﬁon were not

sufficiently precise to allow the fitﬁng of a straight line and its extrapolaﬁon to

36

Dissociation of 3H-0uabain from Myocytes

I—- l—|
.1: m

[.—I
C)

3H-Ouabain Bound (pmollmg pro)
0
oo

 

 

 

 

0.6

0.9

0.2 -

O0 20 (410 60 80 190 120 140 1610 180

Time (min)

Figure 5. Representaﬁve ﬁme course for the dissociaﬁon of 3H-ouabain from
myogytes. Myocytes from guinea-pig heart were incubated in the presence of 0.2
uM H-ouabain for 60 min. Cells were collected by gravity sediigentaﬁon and the
cell pellet suspended in a 100—fold volume of KHB soluﬁon at 37 C containing 0.2
pm non-labeled ouabain. The myocyte suspension was sampled by the rapid
filtraﬁon method at the indicated ﬁmes following iniﬁaﬁon of' dissociaﬁon.
Representa ﬁve pattern of dissociaﬁon from 6 experiments.

37

100Dissociation of 3H-Ouabain from Myocytes

90"
80"\
70r 3‘

o-Total Binding

O-Rapidly Dissociating Binding

0')
O

U1
9’

J:
O
I

N
O
c

3H-Ouabain Bound Z of Time Zero
W
O

 

10

 

o 10 20 30 do 59 60 70
Time (min)

Figure 6. Representaﬁve ﬁme course for the dissociaﬁon of 3H—ouabain from
myocy tes. Time course of dissociaﬁon was determined as described fog Figure 5.
Data is shown on a logarithmic scale as a funcﬁog of the amount of H-ouabain
bound at the iniﬁaﬁon of dissociaﬁon. o -t9tal H-ouabain bound at indicated
ﬁme, o -rapidly dissociaﬁng component of H—ouabgin bound. Representaﬁve
pattern is shown. Dissociaﬁon rate consta_nlt (k ) of rapidly dissociaﬁng
component was calculated to be 0.077501 min (i 1 SE from 6 determinaﬁons)
corresponding to a half-life of 9.0+0.7 min. This figure shows data from a
representaﬁve experiment.

38

ﬁme zero. Despite variaﬁons, the curve after 60 min appeared to be reasonably
ﬂat and therefore, the 60-min value seems to give an approximaﬁon of the X-
intercept. Therefore, the amount of 3H-ouabain remaining bound to cells after 60
min of dissociaﬁon was used as an esﬁmate for the component of bound 3H-

ouabain which is slowly released from myocytes. Subtrac ﬁon of the amount of

3H—ouabain remaining in the cells after 60 min of dissociaﬁon from the amount

remaining at each earlier ﬁme point produced resultant values which had a linear
relaﬁonship when plotted on a semilogarithmic scale (Figure 6). This single
exponential process was described by a dissociaﬁon rate constant (k-l) equal to

1

0.077301 min- corresponding to a half-life of 9.0:0.7 min (mean : SE of 6

experiments). The rate constant for dissociaﬁon of 3H-ouabain from guinea-pig
heart Na,K-ATPase is close to this value (Godfraind e_t gl_., 1980) suggesﬁng the

3

rapid phase of H—ouabain dissociaﬁon from myocytes represents the drug bound

1 as an esﬁmate of the rate of

to the Na,K-ATPase. Using the value of 0.077 min-
3H-ouabain dissociaﬁon it was calculated that after 60 min of dissociation 99% of
drug bound to the rapidly releasing site would have dissociated.

Methods used for esﬁmaﬁng non-specific binding were evaluated to
determine if they produced comparable results in 3H-ouabain binding experiments

3H—ouabain

using myocytes. Nonspecific binding was esﬁmated either as the
remaining bound to cells after 60 min of dissociaﬁon or as the amount bound in
the presence of 1 mM non—labeled ouabain. In this study two methods for
separaﬁng bound from non-bound 3H-ouabain, namely filtraﬁon and centrifuga-
ﬁon, were also evaluated.

Fsﬁmates of total 3H-ouabain binding were not significantly different
when cells were sampled by either filtraﬁon or centrifugaﬁon (Table 1). The

esﬁmates for 3H—ouabain retained by cells after 60 min of dissociaﬁon were also

39

TABLE 1

Esﬁmates of 3H-Ouabain Binding to Myocytes

 

3H—Ouabain Bound (pmol/mg prot)

 

 

Ouabain Slowly
Sampling Method N Total Insensi ﬁve Dissociaﬁng

(1 mM) (60 min)
Filtraﬁon 8 1.11:0.08 0.058:0.008 0.212:0.024*
Centrifugaﬁon 8 1.12:0.05 0.026:0.002* 0.182:0.009*

 

3Myocytes from guinea-pig been} were incubated in the presence of 0.1
uM H-ouabain for 60 min. Total H-ouabain bound was determined after
separaﬁng bound from non-bound ligand by the indicated sampling method.
Non-specific binding was esﬁmated by insluding 1 mM non-labeled ouabain in
the incubaﬁon medium or by allowing H-ouabain to dissociate from the
myocytes for 60 min after binding. 9 Indicates a significant (p< 0.05)
difference in esﬁmates between sampling methods; it indicates significant
gifference between ouabain insensiﬁve and slowly dissociaﬁng components of

H-ouabain binding. Significance determined by Student‘s t-test.

40

not affected by the method of sampling. Nonspecific binding esﬁmated as 3H-

ouabain bound in the presence of 1 mM non-labeled ouabain was significantly
higher when the rapid filtraﬁon method was used to collect the sample and
separate bound from non-bound ligand. More importantly, use of 1 mM ouabain to
obtain esﬁmates of non-Specific binding of 3H-ouabain to myocytes resulted in
values which were significantly lower than those obtained when the dissociaﬁon
method was used (Table 1). This was true regardless of the method used to collect
samples. Because of the extreme morphological change which occurs when
myocytes are incubated with a high concentraﬁon of ouabain, the esﬁmate of
non-specific binding made using the dissociaﬁon method are considered to be
more accurate. These results indicate that esﬁmates of non-specific binding
arrived at by using 1 mM ouabain to block specific binding of 3H-ouabain may
cause the amount of drug bound to Na,K-ATPase to be overeeﬁmated.

Time courses for 3

H-ouabain binding to the rapidly releasing compo-
nent and the slowly releasing component were examined to determine the
opﬁmum incubaﬁon ﬁme for equilibrium binding studies. Myocytes were incu-
bated in KHB soluﬁon containing 50 nM 3H—ouabain for 30 to 120 min. Myocytes
were either sampled at those ﬁmee by filtraﬁon or dispersed for dissociaﬁon of
bound drug and incubated for an addiﬁonal 60 min before being collected. 3H-
ouabain associated with the rapidly releasing site was calculated as the difference
between total binding and 3H—ouabain remaining bound after 60 min of dissocia-

3

ﬁon. H-Ouabain bound to the fast releasing site increased slightly between 30

and 90 min after which it decreased slightly although values were not significantly

different (Figure 7). 3

H—Ouabain associated with the slow releasing component
increased steadily during the 120 min incubaﬁon. These results indicate that a
60-min incubaﬁon is adequate for equilibrium binding studies involving guinea-pig

myocytes when low concentrations of 3H—ouabain are used.

41

Binding of 50 nM 3H-Ouabain to Rapidly or Slowly
0.8’ Dissociating Sites

0
\l
I

Rapidly Dissociating

 

 

 

 

 

 

 

ES
230.6i'
01
E
2:0.5"
0
E
C).
"0.4 P
‘D
C
:3
530.3 -
53
80.2 -
3 Slowly Dissociating * i *
O - a
' - -* 1_.,_1
"$001 +
0 \ l I l I
L" 30 60 90 120

Incubation Time (min)

Figure 7. Binding of 3H-ouabain to rapidly and slowly dissociaﬁng Si s.
Myocytes from guinea-pig heart were incubated in the presence of 50 nM H-
oaubain for the indicateg ﬁme. Myocytes were sampled by rapid filtraﬁon 3for
determinaﬁon of total H-ouabain bound or resuspended to’ allow bound H-
ouabain to dissociate for 60 min. I Indicates specific H—ouabain binding
calculated as th difference between total and that remaining after 60 min of
dissociaﬁon. O H—ouabain remaining bgund after 60 min of dissociaﬁon. *
Indicates a significant (p< .05) change in H-ouabain bound with time of incuba-
ﬁon, n=5. Verﬁcal lines indicate S.E.

42

2. Esﬁmaﬁon of Kineﬁc Parameters for 3 H-Ouabain Bindiggto Guinea-
Pig myogytes

Determination of binding site number and affinity for labeled ligand is
commonly accomplished by analysis of displacement of a labeled ligand from
specific binding sites by various concentraﬁons of non—labeled ligand (Akera and
Cheng, 1977). When this method was used to examine 3H-ouabain binding to
isolated myocytes, a linear relaﬁonship on logarithmic probability paper between

3H-ouabain was

ouabain concentraﬁon and percentage displacement of 100 nM
seen (Figure 8). From this plot a KD = 0.57 uM and Bmax = 10.10 pmol/mg
protein were determined. When an idenﬁcal experiment was performed using
homogenates of cells incubated in 1 mM MgSO4, 1 mM Tris P04 and 20 mM Tris
HCl a KD = 0.17 11M and a B m ax = 9.95 pmol/mg protein were determined. The
linear relaﬁonship on logarithmic probability paper of ouabain concentraﬁon
versus percentage displacement of labeled 3H-ouabain was taken as an indicaﬁon
that the condiﬁons required for using this procedure were satisfied. These are: a
Single class of binding sites and no cooperaﬁvity in binding. The close agreement
in esﬁmates of binding site number for live cells and cell homogenates suggested
that all the 3H-ouabain binding to myocytes was to the Na,K-ATPase.

Linear relaﬁonships on logarithmic probability paper were not always
obtained with this method. For example, when the phorbol ester TPA (phorbol 12-
myristate 13-acetate) was present during 3H—ouabain binding the plot describing
displacement by non-labeled ouabain was clearly non-linear (Figure 8). TPA

sﬁmulates a Ca2

3

+—dependent protein kinase (Nishizuka, 1984). Effects of TPA on
H-ouabain binding were examined to determine if phosphorylaﬁons catalyzed by
this kinase affected the affinity of the Na,K-ATPase for 3H—ouabain. The non-
linearity Shows that the interacﬁon of 3H-ouabain with myocytes is more complex

than originally determined. Therefore, with preparaﬁons of isolated myocytes,

esﬁmates of the dissociaﬁon constant (KD) and binding Site number (Bmax) for

43

 

 

L L L L L

3 190 ' 300

.03 .1 .
Ouabain (uM)

Figure 8. Estimaﬁon of kineﬁc parameters for 3H-ouabain binding by the
displacement method. The presence of 100 nM TPA (pl’tprbol 12-myristate 13-
acetate) during incubaﬁon of myocytes with 100 nil/I H-ouabain caused the
logarithmic probit plot describing displacement of H-ouabain by non-labeled
ouabain to become §on~linear. Percentages are calculated on the mean of six
experiments. B is H-ouabain bound in absence of non-labeled ouabain. D-100
mM TPA; c-coﬁtrol.

44

3H—ouabain cannot be obtained with this method and linearity of plots describing
displacement of 3H-ouabain from myocytes is insufficient evidence for substan-
ﬁaﬁon of these estimates.

More direct esﬁmates of KD and B m x can be made by determining

a
the specific binding in the presence of various concentraﬁons of labeled ligand
and subjecﬁng the data to Scatchard analysis. To uﬁlize Scatchard analysis
correctly it is necessary to expose the cells to a range of ligand concentraﬁons so
that 5 to 95% of binding sites will be occupied. Because full occupancy of binding
sites results in cell death within 60 min, it was necessary to determine the
opﬁmum incubaﬁon ﬁme for a range of 3H—ouabain concentraﬁons. Intracellular
Na+ concentraﬁon is expected to be affected by and to affect 3H-ouabain binding.
Therefore, the ﬁme courses of binding were determined for cells incubated in the
absence and presence of 2.0 uM monensin. Monensin, a Na+ ionophore, is
expected to increase the intracellular Na+ concentraﬁon. The Na+ ionophore was
included to augment the toxic effects of ouabain in order to esﬁmate the maximal
incubaﬁon ﬁme myocytes can tolerate in the presence of high concentraﬁons of
ouabain.

In the absence of monensin, 3H-ouabain binding to the rapidly dissocia-
ﬁng site reached a steady state within 20 min when cells were incubated in 0.5 or

3

2.0 uM H—ouabain but apparently did not reach a steady state within 60 min

3H-ouabain (Figure 9). Ouabain at a

when cells were incubated in 10 11M
concentraﬁon of 0.5 uM has a strong posiﬁve inotropic effect on isolated guinea-
pig cardiac muscle whereas 2.0 and 10.0 uM ouabain are toxic. Decreased Na+
inﬂux into quiescent cells as compared to contracﬁng ﬁssue should ameliorate the
toxicity of ouabain. Therefore, 2 uM ouabain may not be toxic in the isolated

myocyte preparaﬁon and failure to achieve steady-state in binding of 10 uM 3H-

ouabain within 60 min may reﬂect the toxicity of this high concentraﬁon of

45

Time Course of 3H-Ouabain Binding to Rapidly Dissociating Site
25 '

N
O

l—'
U1

A A 10 all 3H-Ouabain
o o 2 uM 3H-Ouabain

 

3H-Ouabain Bound (pmol/mg pro)

 

 

10'
O
. ' ' <> 0 0.5m 3H-0uabain
O
__.o, .- _ -
5 _ . o 9 . O 9“ ,) _“Q
0 . . . A . .
0 10 20 30 40 50 60
Time (min)

Figure 9. Time course of 3H-ouabain binding to the rapidly dissociaﬁng site in
myocytes from guinea-pig heart. Myocytes weﬁe incubated for indicated ﬁmes in:
A, A 10 11M; 0, O 2.0 uM; orO , O 0.5 uM H—ouabain. Open symbols indicate
that 2.0 uM monensin was present during the binding reac ﬁon. n=3. Verﬁcal bars
indicate SE.

46

ouabain. Alternatively, the inability to reach equilibrium may indicate a

redistribuﬁon of Na+ and K+ has occurred during the binding which is of sufficient

magnitude to effect 3

H-ouabain binding. If this is true, 60 min may be
insufficient for both binding and ion redistribuﬁon to approach steady state.

Monensin at a concentration of 2 uM is not itself toxic to cardiac
muscle as judged by a sustained posi ﬁve inotropic response without contracture in
guinea-pig atrial muscle exposed to 1 or 3 uM monensin (Yamamoto gt 9, 1980).
This same concentraﬁon of monensin (2 11M) increases the arrhythmogenic acﬁons
of digoxin (Kennedy gt g” 1983); therefore, it exacerbates the toxicity of cardiac
glycosides.

3

In the presence of 2 uM monensin, binding of H-ouabain is increased

3

early in the reacﬁon at each concentraﬁon of H—ouabain examined. After 20

min of incubaﬁon in the presence of both ouabain and monensin, an apparent

3

release of H—ouabain from cells occurs. For cells incubated in monensin and 2.0

uM 3H-ouabain, the release is sufficient so that the amount retained by the
myocytes is less than the amount bound to cells incubated in 2.0 uM 3H-ouabain
alone. In the presence of monensin, however, binding of 3H-ouabain to the rapidly
releasing site appears to be at steady state after 60 min of incubaﬁon at each
concentraﬁon of 3H—ouabain examined. These data indicate that high concentra-
ﬁons of 3H-ouabain have effects on myocytes, possibly related to an increase in
intracellular Na+, which cause the ﬁme course to equilibrium for binding to
deviate from what is expected for a pseudo first-order reac ﬁon. These data also
indicate that under most condiﬁons, a steady state for the binding reacﬁon is
approached after sixty minutes of incubaﬁon.

2+

Redistribuﬁon of ions, parﬁcularly Na+ and Ca , has been shown to

be the primary result of Na,K-ATPase inhibiﬁon in intact cells. Increases in the

2+

intracellular concentraﬁons of Na+ or Ca may be responsible for the complex

47

behavior of 3H—ouabain binding to myocytes. This could be via a direct acﬁon,
such as Na+ interacﬁng with the Na,K-ATPase, or an indirect acﬁon, such as via a
Ca2+ binding protein or inhibiﬁon of ATP synthesis.

To determine if Na+ or Ca2+ affect binding of cardiac glycosides to
the Na,K-ATPase in intact tissue, myocytes were incubated with concentraﬁons
of 3H-ouabain ranging from 50 nM to 20 11M and the kineﬁc parameters for
binding were esﬁmated from Scatchard plots of specific binding. Specific binding
was esﬁmated as the rapidly dissociaﬁng component of 3H—ouabain binding. The
slowly dissociating component of bound 3H—ouabain was considered to represent
non-Na,K-ATPase binding and, for these experiments, is considered to represent
non—specific binding. Four incubaﬁon condiﬁons were chosen. First, for a control
condiﬁon, cells were incubated in KHB soluﬁon containing 1.8 mM Ca2+. To
augment any effect 3H—ouabain might have via increases in intracellular Na+ and
Ca2+ concentraﬁons myocytes were incubated in KHB soluﬁon containing 1.8 mM

2

Ca + and 2 uM monensin. The third incubaﬁon condiﬁon was chosen to eliminate

effects of Ca2+ without affecﬁng the intracellular Na+ concentraﬁon. For this,
cells were incubated in KHB soluﬁon containing 0.25 mM EGTA (ethyleneglycol

bis(B-aminoethylether)—N,N‘-tetraaceﬁc acid) to chelate Ca2+. Finally, to deter-

mine if increasing intracellular Na+ affected 3

2+, cells were incubated in KHB soluﬁon containing 0.25 mM EGTA to

2

H-ouabain binding in the absence of
Ca
eliminate Ca + and 2.0 uM monensin to increase Na+ inﬂux.

Total and non-specific binding of 3H-ouabain to guinea-pig myocytes
under the four condiﬁons previously described are shown in Figure 10 for
concentraﬁons of 3H—ouabain from 1.0 to 20.0 uM and in Figure 11, on an
expanded scale, for concentraﬁons of 3H-ouabain from 50 nM to 3.0 uM. Binding

3

to the slowly dissociaﬁng site increases with the concentraﬁon of H—ouabain in

the incubation medium under each incubaﬁon condiﬁon. There is a significant

48

Binding of 3H-Ouabain to Myocytes

 

 

 

 

35 " Total 60 min
DISSOC.
32 . e O - C0 n=1-l
0 0 - CC] + M n=Ll i -
A I A - EGTA n=3
228- o ,A - EGTA + M n=li
o» CCI-l.8 mM
€24 " EGTA-0.25 mM *
E s
320-
E- It
§16 - 1'
2/
E. * s- 819. of Ca
012 P . o *
n * -x-
g i" *- SIQ. 01° Mon
9 8- * *
MI i . °
Li #- . /. ’4
E o it
‘I’ ‘0‘
0 1* I 1* I l I
0 l 3 5 10 20

[3H-0uabain] (uM)

Figure 10. Concentra ﬁon dependence for 3H-ouabain binding to myocytes.
Myocytes from guinea-pig hegrt were incubated for 60 min in the presence of the
indicated concentraﬁons of3 l-I-ouabain. Symbols represent C1 , I , O , 0 total
binding or A ,A , O , O H-ouabain remaining bound ﬁtter 60 min of
dissoci 'on. Incubaﬁon soluﬁons contained: 0 , O 1.8 mM Ca (n=4);0,0 1.8
mM Ca and 2 uM monensin (n=4);l , A 0.25 mM EGTA (n=3); 0 , A 0.25 mM
EngA and 2 um monensin (n=4). in Indicates a significant (p<.05) effect of
Ca , -x- indicates a significant effect of monensin. H-ouabain remaining bound
after 60 min of dissociaﬁon is considered to represent non-specific binding.

49

Binding of 3H-Ouabain to Guinea Pig Myocytes

     

 

15 "Total 60 min
Dissoc. at _
e O -CCJ n=ll
1LT - o 0 -C0 + M
A I ‘ "EGTA n=
2 o A -EGTA +
CI.
e. 12 ' Ca=l.8 mM *
§ EGTA=0.25 mM I
O _ I
g 10 . M-2.0 pm
2 s
s 8- -
23 .,
C *
s 6'
:3
cl) Lt" ., *"' Sign Of MO”
MI

 

‘.
. .n/
A “’4 * *
l

o 0.5 1.0 2.0 3.0
[Bil—Ouabain] (pl‘l)

 

Figure 11. Concentraﬁon dependence for 3H-ouabain binding to myocytes.
Myocyte? from guinea-pig heart were incubated in soluﬁon containing 50, 100 or
200 nM H-ouabain or the concentra ﬁons indicated. Incubaﬁons proceeded fogi 60
min. Symbols represent Cl , I , O , 0 total binding orA ,A , O , 9 H-
ouabain remaining bound aftgg 60 min of dissociaﬁon. 2+Incubaﬁon soluﬁons
contained: O , O 1.8 mM Ca (n=4); 0 , O 1.8 mM Ca and 2 pm monensin
(n=4); I , A 0.25 mM EGTA (n=3); D , A 0.25 mM EGTA and 2 uM monensin
(n=4). * Indicates agsignificant (p< .05) effect of Ca , * indictes asignificant
effect of monensin. H—Ouabain remaining bound after 60 min of dissociaﬁon is
considered to represent non-specific binding.

50

effect of monensin to increase non-specific binding when myocytes are incubated

2+, but not when they are incubated in the absence

3

in the presence of 1.8 mM Ca

2+

of Ca2+. Absence of Ca itself also has a Significant effect to increase H-

ouabain binding to the slow releasing site. Whether the presence of 0.25 mM
EGTA instead of the absence of Ca2+ was necessary for or contributed to this
effect was not tested.

Monensin had a significant effect to increase total binding when

myocytes were incubated with less than 3 uM 3

mM Ca2+. Monensin had no effect on binding when cells were incubated in

H-ouabain in the presence of 1.8
3H-
ouabain concentraﬁons greater than 3.0 uM. When incubaﬁons were done in the
absence of Ca2+ (0.25 mM EGTA) monensin did not have any significant effect to
increase total 3H—ouabain binding. Under these condiﬁons there was an apparent
effect of monensin to sﬁmulate 3H-ouabain binding to cells incubated with the 1.0

and 2.0 uM concentraﬁons of 3

H-ouabain, but the increases were not significant
(Figure 11).

Specific 3H—ouabain binding to myocytes was calculated as the differ-
ence in total 3H-ouabain binding and the 3H—ouabain remaining bound to the
myocytes after 60 min of dissociaﬁon. Specific binding of 3H-ouabain to
myocytes incubated under the four condiﬁons described previously is shown in
Figure 12 for concentraﬁons of 3H—ouabain from 1.0 to 20.0 uM and, on an
expanded scale in Figure 13 for concentraﬁons of 3H—ouabain from 50 nM to 3.0
uM. lncubaﬁng cells in the absence of Ca2+ during the binding reacﬁon has a

significant effect to increase specific 3

H—ouabain binding to myocytes incubated
in each concentraﬁon of 3H-ouabain examined. Monensin has a significant effect
on specific 3H-ouabain binding to myocy tes incubated in the presence of Ca2+ but
not in the absence of Ca2+. In the presence of Ca2+ the effect of monensin is to

increase 3H-ouabain binding to cells incubated in concentraﬁons of 3H-ouabain

51

0 _Speciflc Binding of 3H-Ouoboin to Myocytes

 

 

‘-
’8 a: -
a .
mlSr
E
2 at.
8 *.
3 *' 0*
210- * *
3 *. 0 .*
8 .. Co (1.8 mM) n=4
5'. -* 0- Co + Mon (”2.0 uM) n=ll
g l- EGTA (0.25 ITM) n=3
3 5 b u— EGTA + Mon n=ll
Mz': *" 319- Ofm
*‘Slgl OfM‘
0 l l l l .1 I
0 l 2 3 5 l 20

PH-Ouoboln (um

Figure 12. Specific 3H--ouabain binding to myocytes from guinea-pig heart.
Myocytes were incubated in the presence of the indicated concentration of H-
ouabain for 60 min Specific binding was calculated as the difference between
total binding and H-ouabain remaining bomd after 60 min of {lissociatiom
Incubation solutions contained: 0 1.8 mM Ca (n=4); 0 1.8 mM Ca and 2 uM
monensin (n=4); I 0.25 mM EGTA (n=3); C] 0.25 mM2 GTA and 2 pm monensin
(n=4). i Indicates a significant (p< .05) effect of Ca , * indicates a significant
effect of monensin.

52

Specific Binding of 3H-Ouoboin to Myocytes

 

 

 

14'
o- CG(1.8 mM) n=l-l
o- C0 + Mon n=ll
12_ -— EGTA (0.25 mM) n=3
A a - EGTA + Mon n=u *
O
:5 Mon - 2.0 pm *
€10” * *
o a '
s s- .
U
E '*
8 6' ,,
C:
E
O L‘_ 0*
3 ¥
I. . *- 819- OfCO
m I-
2,. *- $19.01“ Mon
’at-
0 L l L I 1 I
0 0.2 0.5 1.0 2.0 3.0

[3H-0uoboln] mm

Figure 13. Specific 3H—ouabain binding to myocytes from guilsea-pig heart.
Myocytes were incubated in solution containing 50, 100 or 200 nM H-ouabain or
the concentration indicated. Incubations were for 60 min. Specific binding ws
calculated as the difference between total binding and 3H—ouabain remaining
bow after 60 min of digsociation. Incubation solutions contained: 0 1.8 mM
Ca (n=4); 0 1.8 mM Ca and 2 11M monensin (n=4); I 0.25 mM EGTA (n=3); [:1
0.25 mM EC21A and 2 um monensin (n=4). I! Indicates a significant (p< .05)
effect of Ca , * indicates a significant effect of monensin.

53

less than 3.0 uM but to decrease specific 3H-ouabain binding to cells incubated in

concentrations of 3H--ouabain higher than 3.0 11M.

3

Saturation of specific H—ouabain binding sites apparently does not

occur under any of the four incubation conditions even when myocytes are
incubated in 20.0 11M 3H-ouabain. A plateau in specific 3H-ouabain binding to
cells is seen at 2.0 to 3.0 uM concentrations of 3H—ouabain when binding is
stimulated by the presence of monensin, or in the absence of Ca2+ (Figure 13).
This plateau may represent saturation of a high affinity site, probably the Na,K—
ATPase.

Scatchard analysis of specific 3H—ouabain binding to guinea-pig myo—
cytes when binding occurs in the presence of 1.8 mM Ca2+ is shown in Figure 14.
When binding is not stimulated by the presence of monensin, a curved Scatchard
plot is produced. For a normal binding reaction, the concave curve shown in
Figure 14 could indicate negative cooperativity in binding or that the 3H-ouabain
is binding to more than one class of sites which have different affinity. For the
cardiac glycosides, binding to Na,K-ATPase in intact cells could be described by
this curve if there is positive cooperativity in binding (Herzig e_t g” 1985a,b).
This point shall be addressed more fully in the Discussion section.

The presence of 2 11M monensin during 3H—ouabain binding causes a
shift in the curve on Scatchard plots, and appears to increase their degree of
curvature. The increased curvature is maintained only for points describing the
binding at concentrations of 3H—ouabain up to 2.0 uM. At higher concentrations

3

of H-ouabain, a break in the curve is seen then the curve continues upward at 10

3

and 20 um H—ouabain.

Scatchard plots describing the concentration dependence of 3

H—oua-
bain binding to cells incubated in the absence of Ca2+ (0.25 mM EGTA) are shown

in Figure 15. Two distinct linear components of this curve are discernable.

54

Scatchard Plots of 3H—Ouoboin Binding to Myocytes

o-C0(L8nMinw
16 ' o— Co (1.8 mM) + Monensin (2.0 M) n=u

I—' H
N .1:
I I

3H—Ouoboin Bound (pmol/mg pro)
5.:
O

 

 

0 1 a l n I 1 a I 1 1 4
0 2 4 6 8 10 12 14 16 18 20 22
Bound/Free x 106

 

Figure 14. Scatchard plots describing specific 3H-ouabain binding to myocytes
from guinea-pig heart. yocytes were incubated for 60-min in solutions contain-
ing concentrations of H—ouabain2 1ranging from 50 nM t$+20 um. Incubation
solutions contained: 0 1.8 mM Ca (n=4) or O 1.8 mM Ca and 2 um monensin
n=4.

55

Scatchard Plots of 3H-Ouobdin Binding to Myocytes
20'

18"

16

I - EGTA (0.25 mM) n=3
n - EGTA + Monensin (2.0 uM) n=4

14.

r

12

10

3H-Ouabdin Bound (pmollmg pro)

 

 

00 21 A 5 3 10 12 14 16 18 20 22

Bound/Free x 106

Figure 15. Scatchard plots describing specific 3H—ouabain binding to myocytes
from guinea-pig heart. éVIyocy tes were incubated for 60-min in solutions contain-
ing concentrations of H—ouabain ranging from 50 nM to 20 11M. Incubation

solutions contained: I 0.25 mM EGTA (n=3) or, D 0.25 mM EGTA and 2 11M
monensin (n=4).

56

Binding to a high affinity site(s) is described by a single line but a break in this

line is accentuated by the curve describing 3H-ouabain binding in the presence of

3

2 uM monensin. If this break is real, H-ouabain binding can be considered to

occur at three sites: the highest affinity site with estimted K = 0.4 11M and

D

Em ax = 9.8 pmol/mg protein; the second, K

protein; and the mird, K

= 0.7 uM, B = 4.3 pmol/mg

D
= 8.4 pmol/mg protein. Monensin

max

D = 3.4 11M, B

would have a modest ability to stimulate
3

max

3H—ouabain binding to the site with

intermediate affinity but no effect on H-ouabain binding to the site with highest
or lowest affinity. Binding to the high affinity site(s) probably represents binding
to the Na,K-ATPase. If this is true, the break in this portion of the curve may
represent binding to different conformations of the Na,K-ATPase. Association
and dissociation rate constants for 3H-ouabain binding to Na,K—ATPase are known
to be affected by the presence of K+ (Akera and Brody, 1970). Therefore, the
presence of 3.8 mM K+ in the incubation media during binding could have caused
two conformations of the Na,K—ATPase with different association and dissociation

rate constants for 3

H-oaubain binding to be present. The physiological nature of
the low affinity site is unknown. For comparison, Scatchard plots describing 3H-
ouabain binding under all four conditions are shown together in Figure 16.
3. Extracellular K+ and alI-Ouabain Binding

Potassium ion has a well documented effect to inhibit 3H-ouabain
binding to isolated Na,K-ATPase (Akera and Brody, 1970). This inhibition
apparently results because K+ binding to the enzyme causes a conformational
change to a form with lower affinity for cardiac glycosides (Akera e_t §l_., 1973;
Inagaki e_t a_1., 1974; Yoda and Yoda, 1986). Normal intracellular K+ concentration
is approximately 140 mM and the extracellular K+ concentration in our experi-

ments is 3.8 mM. Therefore, redistribution of K+ down its concentration gradient

is expected after inhibition of the sodium-pump. Such a redistribution, with the

57

20 Scatchard Plots of 3H-Ouabdin Binding to Myocytes

0 Co (1.8 mM) n=4

0 Co (1.8 mM) + Monensin (2.0 pM) n=u

I EGTA (0.25 mM) n=3

0 EGTA (0.25 mM) + Monensin (2.0 uM) n=4

|—' |—' i—- l-—'
N .1: 0‘0 00
I l l l

5.:
O
I

3H-0u0bdin Bound (pmol/mg pro)

 

 

I l

0 1 l l l 1 l n I 1
0 2 ﬂ 6 8 10 12 1” 16 18 20 22

Bound/Free x 106

Figure 16. Scatchard plots describing specific 3H—ouabain binding to myocytes
from guinea-pig heart. yocytes were incubated for 60—min in solutions contain-
ing concentrations of H-ouabain ragging from 50 nM to 20 “"2'+ Incubation
solutions contained: 0 1.8 mM Ca (n=4) or O 0.18 mM Ca and 2 um
Enon)ensin (n=4); I 0.25 mM EGTA (n=3); [30.25 mM EGTA and 2 uM monensin
n=4 .

58

extracellular K+ concentration increasing as the sodium-pump is inhibited, might
decrease the apparent affinity of Na,K-ATPase for 3H-ouabain if the increase in
extracellular K+ concentration was of sufficient magnitude. This would be
manifested as a negative cooperativity in 3H-ouabain binding and may be
responsible, in part or in whole, for the curvature of the Scatchard plots

3H-ouabain binding to myocytes in the presence of Ca2+.

describing

To demonstrate that Na,K-ATPase can exist in different binding
conformations in intact myocytes, and that extracellular K+ affects the relative
abundance of each conformation, dissociation of bound drug was monitored after
binding occurred in nominally K+-free solution or in KHB solution containing 10
mM KCI (NaCl reduced to maintain osmolarity). A high concentration of 3H-
ouabain (2.5 uM) was present during the binding reaction to assure adequate
binding of 3H—ouabain to the conformation of Na,K-ATPase with the lower

affinity for ouabain. Total 3

H—ouabain binding was estimated by sampling cell
pellets by the centrifugation method before initiation of dissociation. Dissocia-
tion of 3H—ouabain took place in a 200-fold volume of KHB solution containing 2.5
uM non-labeled ouabain and 10 mM K+. Ouabain and K+ were each included in
the dissociation medium to inhibit rebinding of 3H-ouabain released from the
cells. Considerably more 3H-ouabain was bound to cells before initiation of
dissociation than after 30 sec of dissociation. Myocytes contained 23.2:0.80 or
12.5114 pmol/mg protein of 3H—ouabain (n=4) before initiation of dissociation
when binding had occurred in 0 or 10 mM K+, respectively. After 30 sec of
dissociation cells contained 13.4:0.7 or 7.7:0.2 pmol/mg protein of 3H—ouabain
(n=4) for binding in 0 or 10 mM K+, respectively. Such large differences in
estimates 01' 3H-ouabain bound are unlikely to be artifactual, although this is

possible because collection of cells and separation of bound from unbound ligand

were by different methods (centrifugation and rapid filtration). Alternatively, the

59

rapid dissociation of 3

H—ouabain may be real and represent binding to the very low
affinity site seen on Scatchard plots. Because of the large difference in binding
estimated before and 30 sec after initiation of dissociation, the value for 3H-
ouabain bound after 30 sec of dissociaﬁon was used as time zero for analysis of
dissociation.

Release of 3H—ouabain from myocytes was best described by two
exponential processes with short half-lives and a third exponential process with a
much longer half —life. The slowest phase of dissociation is assumed to represent
non-specific binding of 3H-ouabain to myocytes. This phase of release was
estimated from the rate of 3H—ouabain release from cells from 60 to 150 min
after initiation of dissociation. Half lives for these components of 3H-ouabain
binding of 212127 and 146115 min (n=4) were calculated for cells to which binding
had occurred in KHB solution containing 0 or 10 mM K+, respectively. The
difference is not significant and therefore, no conclusions can be drawn concern-
ing the effects of extracellular K” on dissociation of 3H-ouabain from non-
specific binding sites.

Dissociation of 3H—ouabain from the cells between 60 and 150 min
after initiation of dissociation was described by a straight line when data was

3H-ouabain bound to

plotted on a semilogarithmic scale. The best fit line for
cells after 60 to 150 min of dissociation was determined by the method of least
squares and used to estimate the amount of 3H-ouabain associatedwith the slow
releasing component at earlier times of dissociation. Amounts of 3H-ouabain
associated with the slow releasing component at times of dissociation less than 60
min were estimated for individual experiments.

Dissociation of 3H--ouabain from rapidly releasing components was
analyzed using pooled values from four experiments. Individual values from all

experiments were used for calculation of the best fit straight line by least squares

60

analysis. Values from four experiments were pooled because variation in
individual experiments was large. The rapid phase of release of 3H-ouabain was
best described by two exponential processes irrespec tive of binding conditions
(Figure 17). Individual values for 3Iii-ouabain bound to rapidly releasing sites are
shown in Figure 17. The best fit straight lines were calculated by least squares
analysis of individual values from 14 to 39 minutes of dissociation. When binding
had occurred in K+-free solution this component had a half-life of 9.4 min and

3

contained 4.5:0.2 pmol/mg protein of H-ouabain (n=4). When binding had

occurred in 10 mM K+ solution this component had a half-life of 10.8 min and
contained 3.7:0.1 pmol/mg protein of 3H—ouabain (n=4). Correlation coefficients

of the regression lines were 0.93 and 0.94 for binding in the absence and presence

3

of K+, respectively. Individual values describing the most rapid phase of H-

3H-

ouabain dissociation were calculated by subtracting the estimated content of
ouabain in the slower releasing site at times of dissociation less than 11 min from
total 3H-ouabain in the rapidly releasing site. Estimates of 3H—ouabain bound to
the slower releasing site were taken from the regression line describing dissocia-
tion between 14 and 39 min. The best fit lines determined by least squares
analysis of the estimates for 3H-ouabain bound to the fastest releasing site are
drawn in Figure 17 . When binding had occurred in K+-free solution this
component had a half-life of 2.5 min and contained 6.9:0.3 pmol/mg protein of
3H—ouabain (n=4). When binding had occurred in 10 mM K+ solution this
component had a half-life of 3.5 min and contained 2.9:0.1 pmol/mg protein of
3H-ouabain (n=4). Correlation coefficients of the regression lines were 0.92 and
0.89 for binding in the absence and presence of K+, respectively.

The close agreement in dissociation rate constants for 3

3

H—ouabain
bound to cells under these different conditions indicates that H-ouabain binds to

the same sites under both conditions. In the absence of K+ significantly more 3H-

61

Two Phases of 3H-Ouabain Dissociation from
Na,K-ATPase in Myocytes

 
   
    

O
a
0“ Incubation Condition
‘ a : a 10 [M K+
‘1-0 ’ 9 o K’r Free
I

I
N
O

I
\N
O

I

in [Bound/Total Bound at Time Zeraj

 

l

l 1
0 5 10 15 20 25 30 35 40
Time Of Dissociation (min)

Figure 17. Dissociation of 3H-ouabain from myocgtes. Myocytes from guinea—
pig heart were incubated in the presence of 2.5 uM H-ouabain for 40 min. Cells
were collected by gravity sedimentation and resuspended in a 200-fold volume of
KHB solution at 37 C containing 2.5 uM non—labeled ouabain and 10 mM KCl for
initiation of dissociation. Myocyte suspensions were sampled at various times for
150 min following initiation of dissoci 'on. Regression lines calculated from
values normalized to the amount of H—ouabain bound at the first time of
sampling after initiation of dissociation described linear dissociation with time
between 60 and 150 min after initiation of dissociation. Amounts of H—ouabain
associated with this component were calculated from the regression line and
subtracted from all earlier time points. Data plotted are the resultant values
from 4 experimen§ of each type for binding in KHB solution containing .-0 mM
K or 4-10 mM K . Regression lines were calculated from these values and are
shown for two components of the rapid phase+ of dissociation when binding

occurred in KHB solution containing --- 0 mM K or — 10 mM K . See Table 2
for calculated values from regression analysis.

62

ouabain was bound to the cells after 30 sec of dissociation (Table 2). Also, each

3H—ouabain binds, contained significantly greater amounts of

component to which
3H—ouabain when binding had occurred in the absence of K+. This indicates that
in the absence of K+ the Na,K-ATPase remains in a binding conformation for a
longer time. These data are in agreement with published reports describing
effects of K+ on the Na,K-ATPase (Akera and Brody, 1970; Post e_t g” 1972;
Inagaki g g, 1974; Lindenmayer gt a_1., 1974).

Although each component contained a greater amount of 3H-ouabain
when binding had occurred in the absence of K4, the greatest increase by far was
3

in H—ouabain bound to the site with the fastest dissociation rate constant. This

3H-ouabain from isolated

is in agreement with data describing dissociation of
Na,K-ATPase from guinea pig heart (Godfraind gt a_1., 1980).

The ability of K+ to affect the abundance of specific binding sites
indicates the fast component of dissociation represents binding to the Na,K-
ATPase. More importantly, this indicates it may be possible to alter the apparent
affinity of Na,K-ATPase for ouabain by changing the extracellular K+ concentra-
tion. Two mechanisms may be involved; a change in the amount of time the
enzyme spends in a binding conformation and a change in the relative abundance
of high and low affinity conformations of the enzyme.

The magnitude of change in extracellular K+ concentration caused by
sodium-pump inhibition and the concentration dependence of ouabain to cause K+
redistribution were examined by incubating myocytes in the absence or presence
of ouabain for 60 min, then measuring the K+ activity in the incubation media.
Ouabain concentrations used ranged from 50 nM to 20 nM, as had been used to
determine 3H-ouabain binding for Scatchard analysis, and 0.3 mM to determine

maximum redistribution of K+ related to sodium-pump inhibition. Myocyte

63

TABLE 2

Binding and Dissociation of 3H—éiluabain to Guinea-pig
Myocytes Incubated in 2.5 uM H-Ouabain for 40 min

 

 

+ 30 sec . . 31°F". Fast.
Parameter [K ] Dissociation Non-Specific Spgictiefic Spgictiefic

3H—Ouabain 0 13.4103? 2010.2 4510.2 6.910.3
pmol/mg prot. 10 7.710.2“ 1.010.1 ’ 3.710.1 " 2.910.”
r for re— 0 --- > .95 0.93 0.92
gression line 10 --- > .89 0.94 0.89
t1 2 of dis- 0 --- 212127 9.4 2.5
soéiation (min) 10 --- 146115 10.8 3.5

 

Dissociation of 3H-ouabain v§as monitored as described in the text and
legend to Figure 17 . Values are H-ouabain bound to different components as
determined by analysis of the dissociation time course and descriptions of each
regression line. Non-specific binding was determined by analysis of dissociation
between 60 and 150 min after initiation of dissociation. * Indicates a significant
(p< .05) difference caused by the presence of K during the binding reaction.
Analysis of significance was by Student's t-test.

64

density and incubation conditions were identical to those used in determining 3H—
ouabain binding for Scatchard analysis so changes in K+ concentration should be
comparable between these experiments. Total inhibition of the sodium-pump by
0.3 mM ouabain caused an increase in medium K+ concentration of 0.4910.06 or
0.5210.03 mM (n=3) for myocytes incubated in the presence or absence of 2.0 uM
monensin, respectively. Potassium ion concentration of medium in which cells
had been incubated without ouabain was 4.18 mM, and was not affected by
monensin. Therefore, the increase in medium K+ concentration caused by total
sodium-pump inhibition is only 1296 of the basal concentration. This relatively
modest increase in K+ is unlikely to be responsible for the non-linearity of
Scatchard plots. Further support for this is the shift to the right in the ouabain
concentration response curve for release of K+ which is caused by the presence of
2 11M monensin during the incubation (Figure 18). Monensin would have to shift
the ouabain concentration response curve for K+ release to the left if the
increases in K+ concentration caused by ouabain binding were responsible for the
non-linearity of the Scatchard plots. This is deduced because monensin increases
the degree of curvature of Scatchard plots describing 3H—ouabain binding to
myocytes in the presence of Ca2+ (Figure 14), thus indicating that monensin
amplifies whatever process is causing the curvature of the Scatchard plots. Since
the presence of monensin shifts the ouabain concentration-response curve for
release of K+ to the right, monensin inhibits this effect indicating these two
phenomena are not causally related. These results demonstrate that redistribu-
tion of K+ does occur as a result of the exposure of myocytes to high
concentrations of ouabain. The magnitude of change in K+ concentration,
however, is probably not sufficient to affect ouabain binding under incubation

conditions used in determining 3H-ouabain binding for Scatchard analysis.

65

Ouabain Induces K+ Loss by Myocytes
100'

o 1.8 mm Cu”

. o 1.8 mM 002* and 2 mi Monensin

Max= 0.51i.06 mM K+

Basal= 4.171.04 mM K+ ‘ .

OO
O

07
O

N
O

K+ Redistribution 2 of Maximum
.D
O

‘10f005 .01 .02 .05 1.0 2.0 3 5 10 20
Ouabain Concentration (BM)

 

Figure 18. Redistribution of K+ caused by ouabain. Myocytes from guinea-pig
heart were incubated for 60 min in solution containing the indicated concentration
of+ouabain. Cells we e elleted by centrifugation and supernatants analyzed for
K activity (Orion Na AK activity analyzer). Values represent the percentage of
maximal increase in K activ'ty caused by each concentration of ouabain (n=3).
Maximum redistribution of K was assessed by incubating myocy s in 0.3 mM
ouagain. Symbols indicate incubation conditions: 0 1.8 mM Ca ; 0 1.8 mM
Ca and 2 11M monensin. Values represent mean and SE from 3 determinations.

66

4. Binding of 3 H-Ouabain to Myocytes in the Presence of Agents which

élter Intracellular Ca2+ Concentration or Stimulate Production of
Second Messengers

 

Stimulation of 3H—ouabain binding caused by chelation of Ca2+ with

EGTA is consistent with the hypothesis that Ca2+ has a direct effect on cardiac
glycoside-Na,K-ATPase interactions in intact cardiac muscle. A direct effect is

by a mechanism other than that of altering substrate availability, but may involve

2 2

+ dependent protein kinase or a Ca +-binding protein.

2

aCa

O + O I I 0
Since Ca itself at concentrations normally present in cardiac muscle

cells, does not inhibit Na,K-ATPase (Tobin gt a_1., 1973; Godfraind gt a_1., 1977),
chelation of Ca2+ should have no effect to stimulate enzyme activity. Also, it
has been shown with isolated Na,K-ATPase, that in the presence of Na+ and Mg2+,

2+ does not have a significant effect to reduce the time the enzyme exists in a

2+

Ca

binding conformation (Tobin e_t a_1., 1973). Direct effects of Ca ions themselves

on Na,K—ATPase, therefore, cannot be responsible for the stimulation of 3H-
ouabain binding to myocy tes caused by the absence of Ca2+.

A number of Ca2+-dependent protein kinases and Ca2+—binding pro-
teins exist in the heart. It is possible that a Ca2+-dependent protein present in

intact myocytes inhibits 3H—ouabain binding when cells are incubated in milli-

molar concentrations of Ca2+. Negative cooperativity in 3H-ouabain binding

2

. + . . .
would then be seen as intracellular Ca concentration increased in response to

enzyme inhibition.

Direct effects of agents known to alter cell handling of Ca2+ on 3H-

ouabain binding to Na,K-ATPase were examined by incubating myocytes with
these agents and 0.5 or 2.5 uM 3H—ouabain in a KHB solution containing 1.0 mM

K+. Concentrations of 0.5 and 2.5 11M 3

H-ouabain were chosen because they
correspond roughly to the concentrations causing 5096 or full occupancy, respec-

tively, of high affinity binding sites. The low K+ concentration was chosen so that

67

myocytes would be partially sodium-loaded. Sodium loading and low extracellular
K+ concentration will cause the Na,K-ATPase to exist for a greater time in a
binding conformation. Therefore, indirect effects due to Na+ or K+ redistribution
caused by the agents examined are unlikely to affect 3H-ouabain binding. To
maintain cells in a viable condition during incubation in low K+ solution, Ca2+ was

2+ at a 0.1 mM concentration should still be sufficient to

2

reduced to 0.1 mM. Ca

promote any effects caused by Ca + ac ting at an intracellular site because normal

intracellular Ca2+ concentration in resting cardiac cells is less than 1 uM. Entry

2+ into myocytes incubated in KHB solution containing 1.0 mM K+ and 0.1

mM Ca2+ was evident by spontaneous contractile activity observed when myo-

of Ca

cytes were maintained in this solution for greater than 30 min. The source of

2+ supporting the contractile activity was not identified in these eXperiments.

Ca

Agents examined for direct effects on 3H-ouabain binding were: the
phorbol ester TPA (phorbol 12-myristate 13-acetate); the Ca2+ ionophore A23187;
caffeine; lanthanum (La3+); the B-adrenergic agonist isoproterenol; and the 0:1-
adrenergic agonist phenylephrine. Effects of TPA were examined because of its
ability to stimulate Ca2+-phospholipid—dependent protein kinase (PK-C) (Nishi-
zuka, 1984). This enzyme has been postulated to phosphorylate the Na,K-ATPase
of rabbit erythrocytes (Ling and Sapirstein, 1984) and Friend erythroleukemia
cells (Ling and Cantley, 1984). Protein kinase C activity has been found in
cardiac muscle (Kuo gt 91., 1980) and phorbol esters have been shown to have
specific binding sites on isolated myocytes from rat heart (Limas, 1985). Concen-
trations of TPA tested were 3.0 and 30 nM. These concentrations of TPA have a
modest negative inotropic effect on guinea-pig atrial muscle (unpublished obser-
vation) and approximate the KD reported for binding to receptors in rat heart (3.9

2+

nM). Effects of the Ca ionophore, A23187, were examined to determine if

increasing the Ca2+ concentration, beyond that already present in spontaneously

68

contracting myocytes affects 3H-ouabain binding. Effects of caffeine were
examined because this agent is believed to increase intracellular Ca2+ concentra-
tion by effects on the sarcoplasmic reticulum (Blayney gt g, 197 8), and has been
shown to increase resting tension in mammalian cardiac muscle when present at
millimolar concentrations (Chapman and Leaty, 1975; Eisner and Valdeolmillos,
1984). Also, a related methylxanthine, theophylline, has been shown to inhibit the

3

inotropic response of cardiac muscle to ouabain and to inhibit H-ouabain binding

(Zavecz, 1986). Effects of La3+ were examined because this agent is a potent
82

C + antagonist which is effective in blocking Ca2+ movements across sarcolem-

mal membranes. If the effect of Ca2+ on 3H-ouabain binding is mediated by any

2 3+ should

sarcolemmal Ca + transport mechanisms such as Na/Ca exchange, La
block the effect. Effects of B-adrenergic stimulation by isoproterenol were
examined because of this agent's well documented actions to promote phosphory-
lation of sarcolemmal constituents and increase force of contraction. Effects of
al-adrenergic stimulation by phenylephrine were studied because ail agonists have
been shown to stimulate phosphoinositide hydrolysis in cardiac muscle (Brown e_t
a_1., 1985). Phenylephrine has been shown to be an effective agonist of al-
adrenergic receptors in cardiac muscle (Bruckner gt a_1., 1984). Phosphoinositide
hydrolysis has been shown to result in the release of biologically active polyphos-
phoinositols and diacylglycerol into cells (Berridge, 1984; Berridge and Irvine,
1984). Phosphoinositols have unknown actions in cardiac muscle. Diacylglycerol
is an endogenous stimulus for PK-C activity (Nishizuka, 1984).

Caffeine, at a concentration of 3 mM, significantly inhibited 3H-
ouabain binding to myocy tes (Table 3). In a separate experiment, this concentra-
tion of caffeine also decreased the abundance of rod-shaped cells to 51.011.596

(n=5) from greater than 8096 at initiation of incubations. This apparent toxic

action of high concentrations of caffeine makes interpretation of these data

69

«Namaﬁ

. . . . mw<m ENmZ ZO Gum—922.200

 

 

 

 

3 H a ﬂaming: Samsé gemmsé swampea e 3

3 2. 3.9.85 3.9.35 3.9.85 3.9.35 3.943s o m...

2: H 2: 2. 2: 3 e: m
53$. 58?. fa. <2. 3.550

3 2. ”image Semen” .35is 25mg.” Semis o 3

3 3 2.9.3.... 3.9.55 32:? 5.9.3..“ 5.9.36 e a...

25 25 25 3 as 3 25 9... 25 2. 8580 2 2:
FE Fae”: $3 $3 2:895 osotoo 3320.5.

 

Eomcommos 988m Ho .83 2695 83255 .8 couwbcoocoo

«0 933306.55

you“: 3 :32M 3:03. we ooeommzm o5 5 m3 meow—2 993550 mam wﬁcﬁm Sandsctm

m mamﬁy

m

70

ﬂow 300 00én+N00 E 0.83 <Ewm 53 0300305 man 82:20

.0523 £83?m :95:
3 982280 cabana—05 0005 mo 00030 0:033:me 0 853.50.. M SE as 05. 00 SE m4 mo 00:0m08 05
E 9585 Bangalm 0000080O $52.3 50928:: 3:00am “c0093.“ m030> 603065 00 0.8 M 0:0 00 .8
983055280 .Ecm cm .8.“ 580:0.m 0:0 “:03 “$0065 05 no 0000008 05 5 00002305 0.82100» 8945

 

 

 

 

 

.9.
a.” 3 .3..m$.m «mama... Eamza Saws.» gems.“ 25de w 3
3. 3 .39.»: 3.9.8.“ 2.9.2.“ 8.9.80 3.9.2..“ 8.9.8.” 0. m...
25 £5 2: a..." 2: 3 25 3 25 2
«5.00 55$. 85?. + 3 $3 8550
3 2. 3.83.3 Samzﬁ 36.330 Sauna: 3.9.”33 3.9me..“ 0 3
3 2. 3.9.3.4. 3.9.35 2.9.2.6 3.9.3.0 3.9.3.0 3.9.2.5 a 2
2: S 2: N 2: 2: 2: 2 :25 c 23.
mi mam 0a 0% 03.684 3980

 

 

20025 :33 a 030,—.

71

impossible. All other interventions examined were without significant effect on

3H-ouabain binding (Table 3). In these experiments, effects of the various

interventiom on 3H-ouabain binding were examined while the cells were in a
solution low in K+ (1 mM K+) which favors the Na,K—ATPase existing in a binding
conformation. Therefore, effects on 3H—ouabain binding would probably result
from direct effects of the interventions on the affinity of the Na,K—ATPase for
ouabain rather than from indirect effects related to the conformation of the
enzyme. The interventions examined should increase (phenylephrine, A2318?) or

2

decrease (lanthanum) intracellular Ca + concentration or stimulate protein kinase

3

activity (isoproterenol, TPA). The inability of these agents to affect H-ouabain

binding suggests that the mechanism by which Ca2+ removal stimulates 3H-
ouabain binding to myocy tes is indirect.

Stimulation of 3H-ouabain binding by EGTA was re-examined. In the
same experiment the effects of La3+ and A2318? were also studied again. For
this experiment, the K+ concentration was maintained at 3.8 mM so that cells
would be quiescent and intracellular Na+ concentration low. The concentration of

2+

Ca was 1.8 mM except in the incubation media with EGTA where free Ca2+

concentration was reduced to less than 1 nM. The action of EGTA to stimulate

3H—ouabain binding was seen again in this experiment (Table 3). This was probably

3+3

not a secondary result of eliminating Na/Ca exchange activity because La t

the highest concentration tested (0.3 mM) had no effect on 3H—ouabain binding.

2+ 2+

Also, increasing sarcolemmal permeability to Ca with the Ca ionophore,

A23187, did not affect 3H-ouabain binding. These results once again substantiate

that Ca2

+ has no direct effects on Na,K-ATPase when it is present at concentra-
tions compatible with cell viability.
One further important observation from these experiments is that the

Na,K—ATPase of myocytes incubated in solutions containing 1 mM K+, had an

72

apparent increase in affinity for 3

H—ouabain over that in myocytes incubated in
3.8 mM K+ (Table 3). Increased affinity of the Na,K-ATPase for ouabain when
myocytes are in solution with low K+ can be attributed to the enzyme existing in
the Na+-induced binding conformation a greater amount of time. This results
because K+ is required for the change to a non-binding conformation. Additional-
ly, because Na,K—ATPase turnover is reduced, intracellular Na+ concentration
should increase and stimulate the change to a Na+-induced form of the enzyme.
These together will result in maintenance of the Na,K-ATPase in a binding
conformation. A similar increase in affinity was seen when cells were incubated
in KHB soluﬁon containing 3.8 mM K+ and 0.25 m EGTA (Table 3). The increase
in affinity cannot be attributed to removal of a Carr-dependent inhibition of
binding, because when cells were incubated in KHB with 1 mM K+ and 0.1 mM

2+ 2+

Ca was elevated as

2+

, binding was stimulated even though intracellular Ca

evidenced by spontaneous contraction of the myocy tes. To conclude that Ca

does not directly affect binding, an experiment examining the effect of Ca2+ o

3

n
I-I-ouabain binding to myocyte incubated in medium with a low K+ concentration
must be done. An alternative explanation for the stimulation of 3H—ouabain

2+ is that intracellular Na+ concentration

binding which occurs upon removal of Ca
increases under this incubation condition as it does when myocytes are incubated
in low K+ solutions. This has been demonstrated for myocytes isolated from rat
heart (Hohl g1 al_., 1983). These investigators believe that the redistribution of
Na+ in myocytes incubated in Ca2+-free solution is not mediated by ion channels
because verapamil and tetrodotoxin have no effect to inhibit the increase in
intracellular Na+ concentration. It is possible that the increase in intracellular
Na+ concentration is caused by a decrease in Na/Ca exchange activity or is due to
an increase in sarcolemmal permeability to Na+ which may occur in the absence

86

of Ca2+. Results of ouabain sensitive Rb+ uptake studies, however, do not

73

support the concept that an enhancement of 3H—ouabain binding caused by

2+ results from an increase in intracellular Na+ (see below).

removing Ca
C. Sodium-Pump Activity and Capacity Assessed in Myocytes

The physiological representation of Na,K-ATPase is referred to as the
sodium-pump (Ku gt a_1., 1974). Active counter-transport of Na+ and K)r at the
expense of ATP is accomplished by the sodium-pump. Therefore, sodium—pump
activity can be estimated either by measuring active Na+ efﬂux or K+ influx

across the sarcolemmal membrane. Most investigators use active uptake of

86 +

Rb+ as an estimate of sodium pump activity (Akera and Brody, 1985). Rb
substitutes for K+ as a substrate for the Na,K-ATPase but slows down the
reaction, due to slower dissociation of the enzyme-Rb+ complex than of the
enzyme-K+ complex (Post 53 Q” 1972; Tobin e_t g” 1973). This decrease in
reaction rate is advantageous when the function of the sodium—pump is being
examined, because it allows for more accurate estimation of changes in sodium-

86

pump activity (Akera and Brody, 1985). Uptake of Rb+ by the sodium—pump is

86

assessed by measuring Rb+ uptake in the presence and absence of a high

concentration of ouabain. Since ouabain is a specific inhibitor of the Na,K-

86Rb+ uptake can be used as an estimate of sodium-

ATPase the ouabain-sensitive
pump activity.

Because sodium-pump activity is limited by the intracellular Na+ concentra-
tion, measurement of this activity when intracellular Na+ concentration is low
and stable is really an estimate of Na+ influx rate (Yamamoto _e_t 11., 1979; Akera
e_t a_1., 1981). Direct effects on the Na,K-ATPase to alter its activity must be
examined by determining changes in reserve capacity of the sodium-pump.

Reserve capacity of the sodium-pump exists because the enzyme is substrate-

limited by the low intracellular Na+ concentration. When Na+ is accumulating in

74

cells so that its concentration is increasing, or when the cells are sodium-loaded
to the extent that Na+ concentration is not rate-limiting for Na,K-ATPase
activity, the capacity of the sodium pump can be estimated (Akera e_t_ 21:, 1981;
Akera and Brody, 1985). Under these conditions, direct effects of agents on
function of the Na,K—ATPase can be determined.

Isolated myocytes are a unique preparation of cardiac muscle because
extracellular diffusion barriers are eliminated. This is an important consideration

when making determinations of sodium-pump capacity by measuring the ouabain-

86

sensitive Rb+ uptake, because several minutes are generally required for

sufficient tracer to diffuse evenly into whole tissues and be taken up into the

cells. It is expected that individual ions can cross the sarcolemmal membrane

86

several times during the several minutes it takes to assay Rb+ uptake in beating

86

cardiac muscle, thus, causing estimates of specific Rb+ uptake to be low.

Another problem is that when intact tissues are sodium-loaded, activity of the

sodium-pump may be limited by the availability of the counter-ion (86Rb+).

86

Availability of Rb+ in intact tissues may be limited by diffusion barriers within

the tissue when the sodium—pump is functioning at capacity, also causing

86Rb+ uptake to be low. These problems should be

estimates of specific
minimized in preparations of isolated myocytes.

Additional advantages of the myocyte preparation for assessment of sodium-
pump activity are the ease of separating the cells and their content from tracer
ions in the incubation medium and the relatively short time needed to change the
ionic gradients across the sarcolemmal membrane by changing the incubation

6Rb‘k, the centrifugation method was

required for separating accumulated from free 86m)", This was necessary

medium. When estimating cell content of 8

because when myocytes were sampled by rapid filtration, background radioactive

material retained on filters frequently exceeded total counts in samples of cells.

75

The most serious disadvantage of myocyte preparations for estimation of

86

ouabain-sensitive Rb+ uptake is that high voltages are required to electrically

stimulate cells. Threshold voltage for stimulating contractile activity with square
wave pulses of 4 msec duration was greater than 40 V when using plate electrodes
of pure platinum spaced 1 cm apart. This high threshold for electrical stimulation
probably is due to the high electrical resistance of lipid bilayers as compared to
KHB solution and is not an indication of poor cell quality. These myocytes did
respond to electrical stimulation at normal voltages when impaled directly by the
stimulating electrode.9- High voltage requirements for electrical stimulation of
cells precluded this as a mechanism to increase Na+ influx. Under conditions of
the high current flux, electrolysis was visible at the electrode surfaces and cells
died rapidly.
1. Activity and Capacity of the Sodium-pump

Activity of the Na,K-ATPase in isolated myocytes from guinea-pig

heart was estimated as the difference in uptake of 86Rb+ in the absence and

86

presence of 1 mM ouabain. For determination of non-specific Rb+ uptake

86

myocytes were exposed simultaneously to ouabain and Rb+. This may cause

some overestimation of the non-specific uptake because even with a large excess

of ouabain, inhibition of the sodium-pump requires some time. During that time,

some 86Rb+ will enter the cell via the Na,K-ATPase. The slight overestimation

of non-specific 86Rb+ uptake which will occur using this procedure is an

86

acceptable error. The alternative method to estimate non-specific Rb+ uptake

is to incubate myocytes with a high concentration of ouabain before exposing the
cells to 86Rb+. Extended incubation of myocytes in solutions containing 1 mM

ouabain causes redistribution of Na+, K+, and Ca2+ and may alter membrane

 

EDetermination of the excitability of myocytes was done by Robert W.
Hadley working in the laboratory of Dr. Joseph R. Hume at MSU.

76

86 86

Rb+. Therefore, exposure to Rb+

86

properties which affect passive uptake of

and ouabain was simultaneous when determinations of non-specific Rb+ uptake

were made.

Specific uptake of 86Rb+ into quiescent myocytes was linear with

86

respect to time for at least 15 min when ouabain-sensitive Rb+ uptake was

assayed in KHB solution containing 5.0 mM Rb+ (data not shown). Uptake time

86

was limited to 6 min in all assays of ouabain-sensitive Rb+ uptake and was

reduced to 3 or 2 min when examining the capacity of the sodium-pump. Short

86Rb+ uptake were used to ensure that specific uptake would

incubation times for
be linear with respect to time.

Sodium-pump activity was determined in quiescent cells and in cells
exposed to the Na+ ionophore, monensin. Monensin has been shown to stimulate
sodium-pump activity in contracting and quiescent preparations of cardic muscle
(Yamamoto gt a_1., 1979). Presumably, this is a result of its ability to increase
membrane Na+ conductance. For this experiment, exposure of cells to monensin

86Rb+. Monensin had a concentration-

86

was simultaneous with exposure to
dependent effect to increase ouabain-sensitive Rb+ uptake into myocytes
(Figure 19). This effect apparently peaked at 30 uM monensin when uptake was
estimated in the absence or presence of 1.8 mM Ca2+. Specific 86Rb+ uptake into
quiescent cells or cells incubated in low concentrations of monensin was signifi-
cantly lower when the uptake occurred in the presence of 1.8 mM Ca2+ (Figure
19). At higher concentrations of monensin Ca2+ did not have a significant effect

on ouabain-sensitive 86

86

Rb+ uptake. The plateau in the effect of monensin to
stimulate Rb+ uptake into myocytes may indicate that Na+ influx has been
increased to an extent that intracellular Na+ concentration is not rate-limiting

4.
for sodium-pump activity. Alternatively, the ability of monensin to increase Na

77

Monensin Stimulates Sodium Pump Function in Myocytes

   
   

 

 

350—
o- EGTA 0.25mM
300- 2+

3‘3 0- C0 1.8mM
4.:
gﬁzsor *-Sig.of00‘
.Q C
mn—o
38 55 q
(”3200"
> O
..—. L.
4.» o.
"c; 01150”
85
(Dr—0
: “a”
- 2100”
3" o”
o * n 5
s 50 0”

“n=5

0 if“? 3 lb 26 3b 50

[Monensin] mm

Figure 19. Monensin stimulates sodium-pump activity in myocytes. Myocytes
from guinea-pig heart were incubated in the indicated cggcentrations of monensin
ﬂaring incubation for assessment of ouabain-sensitive Rb ugtpke. Uptake of
Rb was measured from solutions containing: 0 1.8 Ca or O 0.25 mM
59135;" Indicates a significant (p< 0.05) effect of Ca on ouabain-sensitive
Rb uptake. Number of determinations as indicated. Significance determined
by Student's t-test.

78

influx may have plateaued. These possibilities were investigated by preincubating
myocytes in monensin for five minutes prior to initiation of 86Rb+ uptake.
In the previous experiment (results shown in Figure 19) exposure to

86Rb+ were simultaneous. Therefore, values represent 86Rb+

monensin and
uptake over a period of time when intracellular Na+ concentration is increasing.
It is possible that by the end of the incubation period (6 min) intracellular Na+
concentration in cells exposed to each concentration of monensin had increased to
the extent that the sodium-pump was operating at capacity. The concentration
dependence of monensin would, therefore, reflect a more rapid increase in
intracellular Na+ at higher concentrations of monensin. Preincubating myocytes

86

with monensin for five minutes before initiating uptake of Rb+ and measuring

86Rb+ caused the maximal effect of

the uptake in KHB solution containing 2 mM
monensin to be reached at a lower concentration (Figure 20). The concentration
of Rb+ was decreased to 2 mM from the 5 mM concentration used in previous
experiments, so that Na+ concentration would increase faster and Na+ would
cease to be the limiting substrate sooner. There was no significant difference in

86Rb+ uptake for cells exposed to 10, 20, 30 or 50 uM monensin when myocytes

were incubated in the absence of Ca2+. A significantly lower 86

Rb uptake was
observed for myocytes incubated in 10 11M monensin in the presence of 1.8 mM
Ca2+, indicating that the effect of monensin was not sufficient to eliminate the
sodium-pump reserve capacity at that concentration. The plateau at 20 to 50 11M
in the effect of monensin to increase ouabain-sensitive 86Rb+ uptake when
myocytes are incubated in a solution containing 2 mM Rb+ suggests that the
sodium-pump is working at its capacity under those conditions. It is unlikely that

the ability of monensin to increase Na+ influx had plateaued, because when uptake

86

of Rb+ was measured from the onset of myocyte exposure to monensin, there

79

Capacity of the Sodium Pump in 2 mM Rb+

110 r DC02+ 1.8 mM
100 ' IEGTA T

9(—

 

 

 

T
1

LO
Q
l

 

i———i

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘ELL—I

 

 

00
O
I

 

 

CT \I

O O

+———4
ft 4

Ouabain Sensitive 86Rb Uptake
(nmol/mg pro/3 min)

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘3333 ——4

 

 

 

 

 

 

 

 

 

 

 

 

 

,T

K Z

f 5

é /

/ é

/ /

50 P ? /

5 5*

r- / /

”0 :3: ;:S

5 ¢

30" ;:: t5:

/ ;

20 L 5 ﬂ
/

¢ ¢

10 b ﬁ f

/ /

é /

0 I &

 

 

 

 

 

10 20 30 50
[Monensin] (11M)

Figure 20. Capacity of the sodium-pump. Ouabain-sensitive 86Rb+ uptake was
measured in myocytes from guinea-pig heart after preincubating the cellssgor g
min in the indicated concentrations of monenzsin. Open bars indicate 86Rb+
mtake occurred in the presence of 1.8 mM Ca , hatched bars indicate Rb
uptake occurred in solution containing 08%5 n1M EGTA. " Indicates a significant
(p< .05) increase in ouabain-sensitive Rb uptake caused by a single step
increase in monensin concentration. n=5 for each determination.

80

were apparent differences (increases) in sodium-pump activity between cells

incubated in 10 or 20 and 20 or 30 11M monensin (Figure 19).

86

The action of Ca2+ to decrease ouabain-sensitive Rb+ uptake was

statistically significant for quiescent cells or cells incubated in low concentra-
tions of monensin but not for cells incubated in high (> 10 uM) concentrations of

monensin (Figures 19 and 20). The concentration dependence for this effect of

2+

Ca was examined using quiescent cells incubated in the absence of monensin.

Concentration of Ca2+

86

was controlled by the presence of 0.25 mM EGTA in the

Rb+ uptake media. Ionized Ca2+ concentration was measured with a Ca2+—

selective electrode and CaCl2 (1.8 M) solution was added to KHB solution
containing 0.25 mM EGTA to increase Ca2+ activity the desired amount. Cells

were incubated in KHB solution containing 1.8 mM Ca2+ prior to initiation of

86Rb+ uptake in solutions with various Ca2+ concentrations. Because the

myocytes were exposed to high or low concentrations of Ca2+

86 2+

only during the

time of Rb+ uptake, effects of extracellular Ca on viability of the myocytes

should be minimal. Increasing Ca2+ concentration from zero to 0.9 mM and from

86

0.9 to 1.8 mM had a significant effect to inhibit ouabain-sensitive Rb+ uptake

(Figure 21). Further increasing Ca2+ concentration to 3.6 and 7.2 mM had no

86Rb+ uptake. If Ca2+ had toxic effects in

the myocytes during the 6-min incubation, a decrease in ouabain sensitive 86Rb+

additional effect on ouabain-sensitive

uptake is expected at the 3.6 and 7.2 mM concentrations of Ca2+. (This, however,

was not observed. This indicates that the action of Ca2+ to inhibit ouabain-

86

sensitive Rb+ uptake is concentration-dependent but in quiescent myocytes the

concentration dependence does not extend to the full range that isolated cardiac

2+

muscle preparations are sensitive to extracellular Ca concentration. In isolated

2

atrial muscle preparations from guinea-pig heart, increasing Ca + has a positive

2+

inotropic effect even when concentrations of Ca exceed 7 mM. This suggests

81

80 C02+ Inhibits Sodium-Pump Activity

70 -
as

so - *

so -

no

30

I

I

I

20

nmol/mg protein/6-m1n

|--‘
O
r

Ouabain Sensitive 86Rb+ Uptake

 

P

L \L L _aL L

1
0 o 0.9 1.8T 3.6 " 7.2
Caz+ Concentration (mM)

Figure 21. Ouabain-sensitive 86Rb+ uptake is inhibited by Ca2+ in a concentra-
tion-dependent manner. Myocy tes from guinea-pig heart wer +added to incuba-
tion media containing the indicated concentrgtions of Ca after a 60-min
incubation in KHB solution containing 1.8 mM Ca '86“1 solutions contained 5 mM
RbCl. Symbols represent: 0 ouabain-sensitive Rb uptake. * Indicates a
significant (p < .05) effect of a one-step increase in Ca concentration (n=4).

82

that the mechanism by which Ca2+ increases force of contraction is different

86

from the mechanism by which it inhibits ouabain-sensitive Rb+ uptake. Inhibi-

86Rb+ uptake by Ca2+

tion of could be due to a direct ac ﬁon on the Na,K-
ATPase, competition for intracellular Na+ via the Na/Ca exchange mechanism,
inhibition of Na+ entry by occupying membrane cation binding sites, or other
mechanisms not considered. The present results cannot distinguish between these
possibilities.

Monensin stimulated sodium-pump activity in myocytes. Because
monersin can have effects in addition to its effect to increase Na+ inﬂux, it was
desirable to demonstrate a stimulation of sodium-pump activity by using another
intervention to increase intracellular Na+ concentration. An alternative method
to stimulate sodium-pump activity in myocytes is to load the cells with Na+ prior

f 86Rb+ uptake (Eisner and Lederer, 1979; Akera and Brody, 1985).

to initiation o
Myocytes were loaded with Na+ by incubating cells in Kt and Rb+-free KHB
solution at 37°C. In the absence of a counter-ion to support Na+ efflux by the
sodium-pump Na+ will accumulate in the cells. Incubation of myocytes with [(+—
and Rb+—free solution was done in a glass column with continuous flow of fresh
solution over the cells to prevent accumulation of K+ as it leaked from the cells.
The Ca2+ concentration of KHB solution used to sodium load the myocytes was
reduced to 10 uM in an attempt to limit toxic effects of Ca2+. After
approximately 30 minutes of perfusion with K+- and Rb+-free solution the
myocytes were all spontaneously contracting. The contracﬁons were not discrete,
however, but resembled a writhing movement. Upon addition of these sodium-

2+ and 3.8 mM K+, the

loaded myocytes to KHB solution containing 1.8 mM Ca
cells began to contract with discrete contractions which included full relaxation
when examined microscopically. Contractions continued for several minutes

following resuspension of sodium-loaded myocytes in solution containing K+ and

83

Ca2+. The sodium-loading procedure did decrease the percentage of cells which

retained a rod-shape in solutions containing Ca2+. This effect was not quanti-

86

tated, however. Measurement of ouabain-sensitive Rb+ uptake in sodium-

loaded myocytes was limited to two minutes in order to minimize toxic effects of

2+ and to limit the decrease in intracellular Na+ concentration which is

Ca
expected to occur when the sodium—pump is activated.
Incubation of myocytes in K+- and Rb+-free solution exhibited a time-

86Rb+ uptake measured when

dependent effect to increase the ouabain-sensitive
cells were exposed to KHB solution containing 5 mM Rb+ (Figure 22). Maximum
stimulation of 86Rb+ uptake was achieved by a 45-min incubation of myocytes in
K+- and Rb+-free solution. Longer incubation of cells, up to 90 min total time,
had no additional effect on sodium—pump activity. The plateau in the effect of
time on sodium-loading could be due to the intracellular Na+ concentration
reaching a plateau or to the sodium-pump operating at capacity. Sodium-pump
activity in sodium-loaded cells is significantly inhibited by the presence of Ca2+
in the uptake medium, similar to the effect of Ca2+ observed in monensin-treated
myocytes. The ability of sodium-loading to stimulate sodium-pump activity
achieves a plateau with the same time for sodium-loading irregardless of the

2+. If the mechanism by which Ca2+ inhibits sodium-

absence or presence of Ca
pump activity is by supporting Na/Ca exchange activity and thereby supporting
competition between the Na,K-ATPase and the Na/Ca exchanger for intracellular
Na+, then intracellular Na+ concentration must have reached a plateau after 45
min of sodium-loading. Alternatively, if the sodium-pump is operating at
capacity, Na+ could still be accumulating in the cells and Ca2+ may act by
directly inhibiting the Na,K-ATPase thus causing the capacity of the sodium-pump

to be reduced when Ca2+ is present (Figure 22).

84

Na+ Loading Stimulates Sodium-Pump Activity

 

 

180 r-
160“ .
O
O O
140 - '
C
s * * .* .*
$120 - f
373
2100 " .* oosau 0.25 mM EGTA
g so _ oosnu 1.8 m C02+
g; , A 86Rb+ Uptake 1 mm Cu. EGTA
‘3 86 + 2+
3 60 L o O Rb Uptake 1 m“! Ca. Ca
53 96 Significant Effect of C02+
:>
to (“3”.
mm .
m 20 h-
a: a: 3‘ 1*
’—
0 lg L L L L L . J
0 15 30 45 50 ‘ 75 90

Time of Sodium Loading (min)

Figure 2 . Stimulation of sodium-pump activity by incubation of myocytes in K+-
fr e, Rb -fr e solution. Myocytes from guinea-pig heﬁ't were superfused with
K -free, Rb -free KHB solution containing 10 uM Ca for Que ingicated time.
Sodium-pump activity was estimated by the ouabain-sensitive Rb uptake over
2 min when 38113 were dispersed in solution containing 5 mM R186 wi+th tracer
amounts of Rb . Open gymbpls represent ouabain-sensitive Rb uptake.
Closed symbols represent Rb uptake in the presence of 1 mM ouabain.
Incubation solutions contained 0, C 1.8 mM Ca (9:5) or,<> , l 0.25 mM EGTA
(n=5). * Indicates a significant (p< .05) effect of Ca

85

86

Non-specific Rb+ uptake was decreased significantly following 15

min of sodium-loading (Figure 22). This may be related to the Na,K—ATPase being

in a high affinity conformation when the cell is sodium-loaded, so that binding of

86Rb+ uptake is accelerated. If cells are

86

ouabain and inhibition of specific

Rb+ uptake is not affected
2+

dispersed into Ca2+-free solution, non-specific
further by increase in the time of sodium-loading. Apparent toxic effects of Ca

which would compromise the sarcolemmal membrane integrity may have been

86Rb+ uptake into cells dispersed into

86

detected as an increase in non-specific

KHB solution containing 1.8 mM Ca2+ (Figure 22). Non-specific Rb+ uptake

increased significantly when myocy tes were exposed to KHB solution devoid of K+

and Rb+ for 45 min or longer. Alternatively, this may reflect an increase in Rb+

2+

conductance of the membrane caused by Ca entry via the Na/Ca exchange

mechanism and activation of a Ca2+-dependent K+ channel (Siegelbaum and Tsien,

1980). If the latter explanation is correct, the increase in non-specific 86Rb+

86

uptake after ouabain-sensitive Rb+ uptake had reached a plateau (Figure 22)

may indicate that the Na+ concentration continues to increase beyond the time

86Rb+ uptake. It is

86

required for the maximum increase in ouabain-sensitive
unlikely that a difference in cell viability would affect ouabain-sensitive Rb+
uptake within a two-minute period or that changing permeability of the membrane

86Rb+ uptake into sodium-loaded cells.

to Na+ would affect ouabain sensitive

Definitive substantiation that the sodium-pump is working at capacity
in sodium-loaded or monensin-treated myocytes would require a demonstration of
an increase in sodium-pump activity resulting from an increase in the number of
pump units or an increase in the intracellular Na+ concentration without an

86

increase in the ouabain-sensitive Rb+ uptake. An attempt was made to increase

the number of sodium-pump units per cell by depleting animals of K+. The

number of 3H—ouabain binding sites in Na,K-ATPase preparations from cardiac

86

muscle of hypokalemic animals has been shown to be significantly greater than
the number of binding sites in Na,K-ATPase preparations from cardiac muscle of
animals with normal concentrations of serum K+ (Erdmann e_t g” 1971; Bluschke
e_t a_1., 1976). Specific 3H-ouabain binding site number has also been shown to
increase in intact cultured chick heart cells after 48 hours of incubation in a
medium containing 1 mM K+ (Kim _e_t a1_., 1984) compared to control cultures
which were maintained in a medium containing 4 mM K+. These data suggest that
cardiac muscle cells respond to low K+ concentration by increasing the number of
functional sodium-pump units.

Guinea pigs were maintained on a K+-deficient diet for 14-21 days in
order to lower serum K+ concentration. Blood was collected from two ketamine-
anesthetized guinea pigs and serum analyzed for K+ activity. Serum K+ activity
in the sample from each animal was less than 2 mM, indicating that they had
abnormally low serum K+ concentrations. Myocytes were isolated from control
and K+-deficient animals simultaneously and analyzed for binding site number and
affinity for 3H—ouabain (Akera and Cheng, 1977) and for sodium-pump activity in
quiescent cells or cells exposed to 5 or 30 11M monensin. There were no apparent
morphological differences in myocytes isolated from normal and hypokalemic
animals. As detailed earlier, kinetic parameters of 3H-ouabain binding to
myocytes cannot be determined by the method used. Therefore, the only reliable
binding data are the estimates of total 3H—ouabain binding to the cells incubated

3

in 100 nM H-ouabain. These values are 0.73:0.04 and 1.24:0.06 pmol/mg protein

3

for cells from control and K+-deficient animals, respectively. The increase in H-

ouabain binding suggests that the binding site number is increased when animals
are maintained on a K+-deficient diet. However, no reliable estimate of non-
specific binding was made so this cannot be certain. Estimates of ouabain-

86

sensitive Rb+ uptake are shown in Figure 23. There was no apparent change

87

\O

[J Control
K+-Deficient ..

oo
0

 

 

 

\l

o

T
\
\

O!
o
U

01
O
1

.b
O
U

° \\\\\\\\\\\i‘}i

 

(p
O
1

 

 

\\\\\\\\\\\\\\\\\\\\

\\\ 'i
°\

Ouabain-sensitive 86Rb Uptake (pmole/mg pro/min)

 

 

 

 

 

 

 

 

 

 

T

 

o
01
o
w
o
o

[Monensin] uM

Figure 23. Sodium-pump activity is not increased in myocytes from K+-deficient
animals. Myocytes rom hearts of guinea pigs fed a control diet (open bars) or a
diet deficient in K (stripetisbars) were assessed for sodium-pump activity by
measuring ouabain-sensitive Rb uptake in the absence of monensin or in the
presence of the indicated concentration of monensin. n=7, vertical lines indicate
SE.

88

86Rb+ uptake of quiescent

caused by K+ deficiency in the ouabain-sensitive
myocy tes or those exposed to 5 or 30 11M monensin. If the apparent change in 3H—
ouabain binding site number corresponds to an actual change in functional sodium-
pump number these data suggest that the reserve capacity of the sodium-pump is
not expressed in cells exposed to 30 uM monensin. Therefore, even when

86Rb+ uptake is an estimate of Na+ influx

monensin is present, ouabain sensitive
and is not affected by the number of enzyme units per cell.

2. Reduction of Reserve Capacity by Interventions Increasig Intracellu-

 

lar Na+ Concentration or Na+ Influx

 

Ouabain has a concentration-dependent effect to decrease sodium-
pump activity. Because the sodium-pump has a reserve capacity, the inhibition of
any given percentage of Na,K-ATPase enzymes is expected to have less effect on
sodium—pump activity when Na+ influx is low than when Na+ influx is high, or
when the cell is Na-loaded (Akera and Brody, 1985). The effect of increased Na+
influx on ouabain inhibition of sodium-pump activity was examined by incubating
myocytes for 60 min in various concentrations of ouabain, then adding cells to

media containing the same concentration of ouabain and tracer amounts of 86Rb+.

86

Ouabain-sensitive Rb+ uptake was determined in the absence and presence of 50

O 0 2+ 0 0
uM monensm and in the absence and presence of Ca . Monensm increased the

86

ouabain-sensitive Rb+ uptake 4—fold (Figure 24) and shifted the ouabain

concentration response curve to the left (Figure 25) indicating that the reserve

capacity of the sodium-pump had been decreased by the augmentation of Na+

2 86

inﬂux. The presence of Ca + during Rb+ uptake decreased sodium-pump

activity (Figure 24) and also shifted the ouabain concentration response curve to

86Rb+ uptake was monitored in the presence of monensin, Ca2+

the left. When
inhibited sodium-pump activity (Figure 24) but did not change the ouabain

concentration response curve. The shif t to the left of the ouabain concentration

89

Inhibition of Sodium Pump Function by Ouabain
350r

o 0 Co” 1.8 mM(n=5)
300 ~ 0 o EGTA 0.25 mM(n=3)

is.
250-% .

N

O

O
l

3...:

U1

0
I

H

O

O
I

 

Ouabain Sensitive 86Rb Uptake
(nmoi/mg pro/6 min)

5%

 

0
O
O
+ ’
l l_

F‘Bul 0:3 . 1:0 3:0
[Ouabain] (uM)

 

Figure 24. Inhibition of sodium-pump activity by ouabain. Myocytes from
guinea-gig lieart were incubated in the indicated concentration of ouabain for 60
min. Rb uptake was initiated by adding myocytes toséolution containing an
identical concentration of ouabain plus grace; amounts of Rb . Closed symbols
represent values for ouabain-sensitive Rb uptake in the absence of mo in
open symbols in the presence of2§0 uM monensin. Incubation media for Rb

uptake contained: 0,913 mM Ca (n=5) or, O , O 0.25 mM EGTA (n=3).

90

Inhibition of Sodium Pump Function by Ouabain

100

so . '\

    
   

60 '
Mon. Control

0 0 Ca 1.8 mM(n=5)
no _ o . EGTA 0.25'mM(n=-3)

1

20

(o

 

l L

0.1 0.13 1.0 3.0
[Ouabain] (um

Ouabain Sensitive 86Rb Uptake Z of Control

Figure 25. Concentration-response curve for ouabain inhibition of sodium-pump
activity. Incubation of myocytes was as described in the legend 85° Figure 2‘}.
Open symbols indicate the presence of 50 11M Hhonensin during Rb uptake,
closed sym ls its absence. Incubation media for Rb uptake contained: 0 ,O
1.8 mM Ca (n=5) or, O, O 0.25 mM EGTA (n=3).

91

response curve caused by the presence of Ca2+ during 86Rb+ uptake in non-

2 86itb+ uptake by

treated cells is inconsistent with an effect of Ca + to inhibit
competition for Na+ via the Na/Ca exchange mechanism. A shift towards the left
could be caused by an increase in ouabain binding or an increase in Na+ inﬂux.
Exposure to a Ca2+-free solution or one containing monensin lasted for 6 min in
experiments to determine ouabain concentration-response curves. Although the

2+ has been shown to stimulate 3

absence of Ca H—ouabain binding to myocytes
after a (50—min incubation (Figure 15), the time course for this phenomena was not
examined. Monensin significantly increased 3H—ouabain binding after 5 min of
exposure (Figure 9). Since the specific binding of ouabain was not examined in
this experiment, the shift in the ouabain concentration response curve does not
unequivocally demonstrate a decrease in reserve capacity. The shift in the
ouabain concenb‘ation-response curve caused by the presence of Ca2+ is inconsis-
tent with the effect of Ca2+ to decrease 3H-ouabain binding but is consistent with
a decrease in reserve capacity. The greater shift seen under conditions of

3

increased Na+ inﬂux is consistent with both an increase in H-ouabain binding and

a decrease in the reserve capacity of the sodium-pump caused by monensin.

DISCUSSION

A. R_egt_ilation of the Na,K-ATPase and Cardiac Glycoside Binding in Intact
Tissue

 

Positive inotropic effects of the cardiac glycosides on cardiac muscle and
the ability of these agents to bind to specific receptor sites on the Na,K-ATPase
causing inhibition of the enzyme have each been studied extensively (Lee and
Klaus, 1971; Akera and Brody, 1978). Overwhelming evidence exists which
supports the hypothesis that inhibition of the Na,K-ATPase is causally linked to
both the positive inotropic effect and the toxic effects of the cardiac glycosides.
One major question concerning the Na,K-ATPase which is still unanswered is
whether or not there are endogenous mechanisms for regulation of the enzyme
activity (Anner, 1985). In attempts to answer this question an extensive amount
of work has been done, but at this time the only factor known to affect activity of
the Na,K-ATPase is substrate availability -- specifically, intracellular Na+
concentration (Akera and Brody, 1982).

In order to demonstrate that an intervention or regulatory factor has an
effect on Na,K-ATPase activity, the enzyme function must be examined under
conditions such that the Na+ concentration is not rate-limiting. This condition
probably occurs transiently with each contraction of the heart, but at this time
the technology is not available to monitor Na+ transients; therefore, there is no
way to directly monitor this effect in intact tissue.

The isolated myocyte preparation has many features which may facilitate

estimation of drug binding and sodium-pump activity. These parameters have

92

93

been examined in isolated myocytes from guinea-pig heart and indicate direct
effects of agents on Na,K-ATPase activity and cardiac glycoside binding can be
studied with this preparation.

B. Myocyte Isolation for Examination of 3

Ac tivity
Myocytes from guinea-pig heart were isolated in high yield and with greater

H-Ouabain Binding and Sodium-Pump

than 8096 of all cells retaining normal morphology. These latter cells were

quiescent in solutions containing millimolar concentrations of Ca2+

and possessed
normal characterisb'cs of excitable tissues in terms of membrane electrical
properties. The properties of these cells and their stability over time compare
favorably with myocyte preparations isolated by other investigators which have
been described in the literature (Farmer gt $1., 1977; Haworth e_t al_., 1980, 1982;
Isenberg and Klockner, 1982; Bihler e_t a_1., 1984; Bkaily e_t g” 1984). Differences
in isolation procedures which promote isolation of stable cells were not tested
directly. Therefore, a comparison of the method for myocyte isolation described
in this work with other published methods is not possible. A review and
comparison of published procedures for isolation of cardiac myocytes has been
published (Farmer e_t a_1., 1983).

From the work done in this laboratory, two points seem very clear; first, to
isolate myocytes in high yield, hearts must be extremely well digested; second,
isolated myocytes are very fragile, and to maintain these cells in viable condition
they must be treated gently. The second point may explain the need for good
digestion of the tissue; as a well digested tissue requires a minimum of
mechanical agitation to dissociate the cells. Collagenase and hyaluronidase were
used to digest the connective tissues in the hearts. Collagenase was extremely

critical for this process and variation in this enzyme was responsible for most

variation in the isolation procedure. Hyaluronidase is not essential for the

94

digestion, and on one occasion this enzyme was not used. Omission of hyaluroni-
dase did not appear to affect digestion of the tissue or morphology of the cells but
this enzyme was included so that there was consistency in the isolation procedure
throughout the entire course of these experiments.

The time course of the digestion procedure is well detailed in the Methods
section. It seemed important to follow this to as great an extent as possible. The
changes in digestion procedure which were used are: increase in time of exposure
to enzymes, decrease in time of perfusion wiﬂi low Ca2+ medium before starting
recirculation, and decrease in volume of recirculating perfusate. Increasing the
time of exposure to enzymes seems an obvious way to enhance digestion.

Another consideration, however, is that the enzymes must be in a low Ca2+

2

solution to be effective (increasing Ca + concentration to even 50 uM in the

solution containing enzymes inhibited digestion) and cardiac muscle is subject to a

2 2+ paradox"

phenomenon termed the "Ca + paradox" (Baker e_t a_1., 1983). The "Ca
is a toxic effect of physiologic concentrations of Ca2+ (millimolar) after the heart
muscle has been exposed to a Ca2+-free medium. This phenomenon may be the
reason it is difficult to isolate Ca2+-tolerant myocytes from cardiac muscle. To

2

minimize the extent of the "Ca + paradox" during isolation, Ca2+ (10 nM) was

added to all buffers and the amount of time the tissue was perfused with low
Ca2+ KHB solution was minimized. It is not obvious why reducing the time of
perfusion with low Ca2+ medium before starting recirculation or reducing the
volume of recirculating medium should enhance digestion. These observations
were made consistently, however. A possible explanation for the effectiveness of
these changes is the release of some factor from the heart during the perfusion

2+ KHB solution which enhances digestion. An observation which is

with low Ca
consistent with this is that large hearts were digested more successfully than

small hearts. Incorporating these variations into the normal digestion procedure

95

2+ . .
media before recircu-

was not done because insufficient perfusion with low Ca
lation was initiated caused visible tissue damage. This was evident from white
areas on the heart. Volume of recirculating perfusate was not reduced because
smaller volumes of perfusion medium were feared to be inadequate for maintain-
ing ﬂie tissue for the required 52 min of perfusion.

Selection of rod-shaped cells was possible because these cells formed loose
clusters more readily than rounded cells and because the latter apparently have a
greater surface tension. The greater surface tension of rounded cells is a
deduction based on the observation that in a droplet of medium containing cells,
only rounded cells are on the surface exposed to air whereas both rod-shaped and
rounded cells are on the glass surface of the microscope slide. Aggregates of
cells have a faster sedimentation rate in the KHB solution than individual cells.
This is true irrespective of the shape of the individual cells or the composition of
the cell aggregates with regard to cell morphology. The tendency for rod-shaped
cells to form aggregates made it possible to elutriate these cells by gravity
against a ﬂow of KHB solution. Elutriation of these preparations is an ideal way
to prevent the myocytes from forming packed pellets which are relatively
inaccessible to the oxygen and nutrients dissolved in solution. This step in the
isolation procedure seems to be a significant technical advance for maintaining
myocytes in a viable condition after their dissociation. It also significantly
increases the percentage of rod-shaped cells in the preparations because single
cells are not retained in the column used for the elutriation procedure and
rounded cells are less likely to form aggregates.

Final selection for rod-shaped cells was by gravity sedimentation in a test
tube. The higher surface tension of the rounded cells made it possible to aspirate
media in which myocytes and myocyte aggregates had only partially sedimented

and to preferentially remove rounded cells. This was done by aspirating only the

96

surface of the media from one edge so that a circular ﬂow of medium within the
tube was created. Aggregates sedimented more quickly and were less likely to be
caught in the current of medium. When they were caught, aggregates were less
likely to be aspirated out of solution than the single cells which had higher surface
tension.

Significant enhancement of myocyte viability with time by ouabain at a
concentration (1 uM) which is toxic to guinea-pig cardiac muscle contracting at a
normal frequency was very surprising. This may indicate that energy expenditure
to maintain ion gradients is high in the isolated myocytes. No determinations of
oxygen consumption by myocy tes or the effect of ouabain on oxygen consumption
were made, however, so this is only speculation. Alternatively, inhibition of the
Na,K-ATPase may stabilize the membrane and promote myocyte viability or 1 uM
ouabain may not be toxic because these cells remained quiescent. Beat- or
depolarization—dependency of the onset of glycoside action is well established.
Toxicity of ouabain at a concentration which caused total inactivation of the

sodium-pump (1 mM) is to be expected.

C. Estimation of 3I-I-Ouabain Binding to Guinea-Pig Cardiac Myocytes

Specific binding of cardiac glycosides to intact cardiac muscle can be
estimated directly (Godfraind and Lesne, 1972; Busse gt a_1., 1979; Herzig gt a_1.,
1985; Kjelden gt gl., 1985) or indirectly (Ku gt a_1., 1974; Gelbart and Goldman,
1977; Temma and Akera, 1982). The indirect estimate measures initial velocity of

binding of a 3

H—glycoside to a sample of tissue homogenate. Initial velocity of
binding is proportional to the free receptor concentration (Gelbart and Goldman,
1977), therefore, the number of receptors which had non-labeled glycoside bound
can be estimated. Determinations of receptor occupancy by this method can

provide information about the relative number of free receptors, but are not

97

useful to obtain data for estimation of kinetic parameters of binding. Direct

measurement of 3H-ouabain or 3

H-digoxin binding to intact muscle has the
disadvantage of high non-specific binding. Whether it is estimated by including an
excess of non-labeled ouabain (Godfraind and Lesne, 1972) or by estimating
incorporation into tissue components by monitoring dissociation after binding
(Kjelden _e_t a_1., 1985) the non-specific binding accounts for approximately 5096 of
total binding. High non-specific binding makes accurate determination of specific
binding difficult. Intact tissues have the additional disadvantage of containing
many cell types; therefore, the percentage of binding to receptors which are on
muscle cells cannot be determined. Despite these problems the concentration

3

dependence of H—ouabain binding has been assessed in guinea pig atrial (Busse e_t

a_1., 1979; Herzig e_t a_1., 1985) and ventricular muscle (Kjelden gt a_1., 1985). Data

3

from these studies are generally consistent with the results of the H-ouabain

binding studies utilizing guinea-pig myocy tes. These results will be considered in

more detail when the concentration dependence of 3

H-ouabain binding to guinea-
pig myocytes is discussed.

Binding of 3H—ouabain to isolated cardiac muscle cells in culture (Friedman
gt a_1., 1980; Werden e_t a_1., 1983; Kim e_t g, 1984) and to dissociated myocytes
from adult animals (Onji and Liu, 1981; Adams gt a_1., 1982) have also been
reported. These investigators find a single class of high affinity binding site in
cells derived from embryonic chick ventricle (Werden gt g” 1983; Kim e_t gl_.,
1984), a single class of high affinity binding site in cells from rat heart (Friedman
e_t 91., 1980; Adams gt a_1., 1982), and two classes of high affinity binding sites for
3H-ouabain in myocytes from dog heart (Onji and Liu, 1980).

In these reports of 3H-ouabain binding to isolated myocytes or cultured cells

no difficulty was reported in determining non-specific binding. Each experimental

design utilized an excess (millimolar) concentration of non-labeled ouabain to

98

‘.

block specific binding of 3H-ouabain when determinations of non-specific binding
were made. Toxic effects of these saturating concentrations of ouabain were not

reported but certainly must have occurred. Toxicity of ouabain may have not

2+

been observed by these investigators because Ca was not present in the reaction

mixture (Friedman gt gl_., 1980; Onji and Liu, 1981; Werden e_t g” 1983) or was
present at a low (50 11M) concentration (Kim gt gl., 1984). A higher (0.5 mM)

concentration of Ca2+

was present during the incubation of myocytes with
ouabain described by Adams and coworkers (1982) but the incubation time was
short (30 min) and the cells examined were from rat heart which contains mainly

the sodium—pump which has low affinity for 3

H-ouabain (Akera e_t g” 1979;
Adams e_t g, 1982). In experiments with myocytes from guinea-pig heart,
exposure to millimolar concentrations of ouabain did not cause spontaneous

2+ until cells had been

contractile activity in KHB solution containing 1.8 mM Ca
incubated with ouabain for 10 to 15 min. Time required to induce contracture of
these myocytes by ouabain varied widely within preparations, but most cells
retained a rod-shape for at least 30 min. Therefore, the toxic reactions of
myocytes to ouabain are slow to develop and may not be significant with short
incubations. Binding reactions were allowed to proceed for 60 or 240 min in
experiments reported by Onji and Liu (1981) or Werden gt a_l. (1983), respectively.
Toxic effects of ouabain may not have developed even with these longer

2+

incubation times due to the absence of Ca . Alternatively, adverse reactions to

ouabain may have been overlooked.

Data from experiments with guinea-pig myocytes clearly demonstrate that
myocytes are not stable when incubated in a combination of ouabain and Ca2+ if
each is present at a millimolar concentration. The effect of 1 mM ouabain to

3

decrease non—specific binding to myocytes incubated with 0.2 uM H-ouabain is

very clear. Whether the toxicity of ouabain in the presence of Ca2+ is responsible

99

for the decrease in non-specific binding is not certain. The possibility exists that
ouabain decreases estimates of non-specific binding by blocking a putative uptake
process for 3H-ouabain. If it exists, this uptake may proceed as a transport of
glycoside by the Na,K-ATPase (Dutta e_t Q” 1968; Park and Vincenzi, 1975; Fricke
and Klaus, 1977) or as an internalization of the glycoside-enzyme complex
(Pollack e_t a_1., 1981; Cook gt a_1., 1982). lnternalization of the ouabain-receptor
complex is proposed to occur as part of the process of Na,K-ATPase turnover and
glycoside binding is proposed to increase the rate of Na,K-ATPase turnover (Aiton
e_t a_1., 1981). The evidence for internalization of the ouabain-receptor complex
comes from experiments with HeLa cells, however, and therefore may not be
applicable to mammalian cardiac myocytes.

Transport of cardiac glycosides into cardiac muscle cells has been proposed
to occur and the intracellular injection of glycosides has been reported to increase
velocity of shortening in isolated myocytes (Isenberg, 1984); thus, suggesting an
intracellular site exists for cardiac glycoside binding. Transport of cardiac
glycosides, however, has not been demonstrated in cardiac muscle. Localization

of3

H-ouabain in the microsomal fraction of cardiac muscle homogenates pre-
pared from tissue exposed to 3H—ouabain prior to homogenization and fractiona—
tion suggested to some investigators that the glycoside was transported into the
cells and then bound to sites in the sarcoplasmic reticulum (Dutta e_t a_1., 1968).
This was taken as evidence that the glycosides had specific binding sites on the
sarcoplasmic reticulum. Because it is virtually impossible to isolate either
sarcolemmal or sarcoplasmic reticulum vesicles which are not contaminated to

3H—ouabain has an

some extent by the other fraction, the conclusion that
intracellular binding site is not justified from these studies. Association of 3H-
glycosides with cardiac muscle which is affected by the Na+ concentration (Fricke

and Klaus, 1977) or the species of glycoside used (Godfraind and Lesne, 1972; Park

100

and Vincenzi, 1975) was used as evidence that the Na,K-ATPase transports
cardiac glycosides into muscle cells. The basis for this conclusion, however, is not
obvious as similar changes in tissue concentrations of glycoside would be expected
if specific binding of cardiac glycosides to extracellular sites is affected by the
Na+ concentration or the species of the glycoside. Since both of these are well

documented to affect 3

H-glycoside binding to Na,K-ATPase, the available evi-
dence does not indicate that the Na,K-ATPase transports glycosides into cardiac
muscle.

Neither mechanism (toxicity or blocking an uptake process), by which non-
specific binding of 3H-ouabain would be decreased in the presence of high
concentrations of non-labeled ouabain is supported by evidence published by other
investigators. Nevertheless, the experimental design used appears sound and
sufficient to detect the reported effect of millimolar ouabain concentrations to
reduce estimates of non-specific 3H—ouabain binding. Therefore, it must be
concluded that including excess non-labeled ouabain in order to make determina-
tions of non-specific 3H-ouabain binding to guinea-pig myocy tes in the presence
of Ca2+ is a method subject to large errors.

Measurement of bound 3H-ouabain before and after partial dissociation at
0°C has been used to estimate specific and non-specific binding to guinea-pig
skeletal and cardiac muscle (Kjeldsen _e_t gl_., 1985). In these experiments specific
binding is quantitated as the slow phase of release. Dissociation of 3H—ouabain
from guinea-pig myocytes at 37°C occurred in fast and slow phases. The fast
phase had two components, the relative abundance of each being dependent on the
presence of K+ during binding, thus indicating that this represents binding to

3H-ouabain bound to

Na,K-ATPase. The dissociation rate constant calculated for
the rapidly releasing site in guinea pig myocytes is similar to the value reported

for the dissociation rate constant of 3H-ouabain bound to the Na,K-ATPase from

101

guinea pig heart (Godfraind gt gl_., 1980), further supporting the hypothesis that
the fast phase of 3H-ouabain release from guinea-pig myocytes represents 3H-
ouabain bound to the Na,K-ATPase. The slow phase is assumed to represent non-
ATPase binding and to be non-specific. A half-life for release of 3H-ouabain from
non-specific binding sites was not determined in initial experiments due to the
large relative errors in sampling and counting and the small amount of 3H—ouabain
associated with this site. Therefore, 3H-ouabain remaining bound after 60 min of
dissociation at 37°C in a 100-fold volume of KHB solution containing 13.8 mM K+
and with a concentration of non-labeled ouabain equal to the concentration of 3H-
ouabain with which the myocytes had been incubated, was used as an estimate of
non-specific binding.

It is unreasonable to expect that non-specific binding is irreversible. In
fact, at 0°C non—specifically bound 3H-ouabain is released more rapidly than is
the specifically bound ligand (Kjeldsen e_t a_l_., 1985). In experiments examining the
effect of K+ on specific 3H—ouabain binding, dissociation of 3H-ouabain was
monitored after binding had proceeded for 40 min in the presence of 2.5 uM 3H-
ouabain and either 0 or 10 mM K+. With this high concentration of 3H—ouabain
present during the binding reaction, sufficient ligand was associated with the
slowly releasing site so that its rate of dissociation could be estimated. Half lives
of approximately 212 and 146 min were calculated for dissociation of non-

specifically bound 3

H-ouabain, when binding had occurred in the absence or
presence of 10 mM K+, respectively. These values are not significantly different,
therefore, no conclusions can be drawn concerning the effects of K+ on the
dissociation of 3H-ouabain from non-specific binding sites. If the faster rate of

release is the more accurate estimate of dissociation from this site, then after 60

min 2596 of non-specifically bound 3H-ouabain would be released from the

102

myocytes. Therefore, if non-specific binding accounted for 20% of total 3H-

ouabain binding, an overestimation of approximately 7% for specific binding would
occur because too small a value for non-specific binding would be subtracted.
Even with this systematic error to underestimate non-specific binding, the
dissociation method produced estimates for this component of binding which were
more than three times as great as estimates made by including 1 mM ouabain in
the incubation mixture to block specific binding.

It is unlikely that such a large difference is due to overestimating the non-

3

specific binding by the dissociation method. Overestimation of the amount of H-

ouabain associated with the slow-releasing site could occur due to allowing
inadequate time for dissociation or insufficient dilution of 3H—ouabain during
dissociation. Because dissociation occurred in KHB solution containing the same

concentration of non—labeled ouabain as the binding incubation solution had

3

contained H-ouabain, no change in specific binding should occur and 3H—ouabain

should occupy 1% of those sites when equilibrium is re—established. Therefore,

the maximum overestimate of non-specific binding due to inadequate dilution is

3

1% of the specific binding. Rebinding of H-ouabain would cause the dissociation

half-life to be overestimated rather than underestimated, so it is unlikely that

insufficient time had been allowed for dissociation. The largest estimate of half-

3

life for H-ouabain dissociation from specific binding sites was 10.8 min. If this is

3

an accurate estimate then 2% of specific binding sites which had H—ouabain

3H-ouabain bound after 60 min

bound at the initiation of dissociation would have
of dissociation. Errors due to inadequate dilution and insufficient time of
dissociation should be roughly additive when both are small. Therefore, the
maximum overestimate of non-specific binding by this method should be 3% of the

specific binding.

103

From the available information concerning cardiac glycoside binding to
myocytes from guinea-pig heart in the presence of Ca2+, it is apparent that the
most accurate estimates of specific binding can be obtained by using the
dissociation method to estimate non-specific binding. Estimating the non-specific
binding of 3H—ouabain as the amount of 3H—ligand remaining bound to the
myocytes after 60 min of dissociation in a 100—fold volume of KHB solution is
subject to systematic errors which can cause non-specific binding to be underesti-
mated by as much as 25% of the non-specific component of binding or overesti-
mated by as much as 3% of the specific component of binding. As the relative
amount of non-specific binding increases, the underestimation of non-specific
binding can become a serious problem. Therefore, the dissociation method may
not be adequate to determine non-specific binding when that binding accounts for
greater than 20% of the total binding.

Estimation of kinetic parameters for drug receptor interactions is generally
accomplished by analysis of displacement from specific binding sites of 3H-ligand
present at a constant concentration in the incubation media, caused by the
presence of various concentrations of non-labeled ligand. Alternatively, specific
binding can be determined when receptors are incubated with various concentra-

3H—ligand. In the first instance displacement of 3H-ligand by non-labeled

tions of
ligand can be analyzed according to the method of Akera and Cheng (1977). These
authors are careful to point out that this method for estimating kinetic para-
meters can only be used when there is a single class of binding site and where
there is no cooperativity in drug binding. When this method was used to examine
3H-ouabain binding to guinea-pig myocytes in the presence of Ca2+ and K+, the
data appeared to confirm that 3H-ouabain binds to a single class of receptors and

that there is no cooperativity. This, however, is apparently not true. The first

evidence indicating this was the non-linearity of the logarithmic probit plots

104

describing displacement of 100 nM 3H—ouabain by non-labeled ouabain when
binding to guinea-pig myocytes occurred in the presence of TPA. Later

3

experiments using Scatchard analysis to estimate kinetic parameters for H-

ouabain binding indicated that there may be more than one binding site with

different affinity for 3

H—ouabain in guinea-pig myocytes and/or that there may
be cooperativity in binding. These data are consistent with findings from
experiments examining 3H—ouabain binding to isolated Na,K-ATPase from guinea-
pig heart in the presence of K+. Under these conditions 3H-ouabain binds to
conformations of the enzyme which have different affinities for the glycoside
(Godfraind gt a_1., 1980). These data are also consistent with the theoretical
considerations concerning glycoside binding to Na,K-ATPase in intact tissue which
predict cooperativity in binding (Herzig gt g_l., 1985a,b). Thus, experimental
evidence and theoretical considerations both indicate that the method of displace-
ment cannot accurately estimate kinetic parameters of cardiac glycoside binding
to Na,K-ATPase in intact myocytes when binding occurs in the presence of Ca2+
and K+. It may be possible, however, to use this method if Ca2+ is not present, or
if K+ is present in very low concentrations. Under these conditions cooperativity
in binding has not been demonstrated and a single class of binding site may
predominate.

3

Estimates of kinetic parameters for H—ouabain binding to high affinity sites

in myocy tes as determined by the displacement method have been published by
Adams and coworkers (1982). Binding reactions to myocytes from rat heart
occurred in an incubation solution containing 3.8 mM K+ and 0.5 mM Ca2+.
Therefore, cooperativity in binding is expected. However, Na,K-ATPase from rat
cardiac muscle is unusual because two forms of the enzyme exist and only about
15% of the total binding sites have a high affinity for ouabain (Adams gt a_1.,

3

1982). Because so few sites have a high affinity for H-ouabain, it may be

105

possible to inactivate those pump sites without significant positive cooperativity
resulting from the increased intracellular Na+ concentration. Thus, determination
of 3H—ouabain binding to cardiac myocytes from rat may not be subject to the
same limitations that exist for guinea-pig myocytes. When a single class of
binding sites are present, or when two classes with a similar affinity are present,
such as with guinea-pig myocy tes, and binding occurs in the presence of Ca2+ and
K+, the displacement method is not adequate for estimation of kinetic parameters
for 3H-ouabain binding.

Scatchard analysis of 3H-ouabain binding to specific binding sites is a more
direct method for estimation of kinetic parameters than is the method of
displacement. Guinea—pig myocytes when incubated in the presence of 1.8 mM

2+ 3H—ouabain, bind the 3H-ouabain in a manner

Ca and various concentrations of
such that Scatchard plots describing the specific binding are non-linear. Other
investigators have found binding to cultured chick heart cells is described by
linear Scatchard plots (Werden _e_t g” 1983, 1984; Kim e_t a_1., 1984). Binding to
dog cardiac myocytes, however, also is described by non-linear plots (Onji and Liu,
1981). Linear Scatchard plots were interpreted to indicate the presence of a
single class of binding sites in cultured chick heart cells. Non-linear Scatchard
plots were interpreted to indicate the presence of two classes of binding sites in
myocytes from dog heart; these sites were shown to be interconvertible by the
addition of K+, and therefore probably represent different conformatiOns of Na,K-
ATPase. The relationship of these findings to the curved Scatchard plot
describing 3H-ouabain binding to guinea-pig myocy tes is uncertain because in each
of the studies 3H—ouabain binding to cells had occurred in the absence of Ca2+ and
in low K+ or K+-free solution. The presence of Ca2+ is a major factor in non-

linearity of Scatchard lots describing 3H—ouabain binding to guinea-pig myocytes.

This is not the only factor, however, because when binding occurs in the absence

106

of Ca2+, the Scatchard plot has at least two components, representing high and
low affinity binding sites.
Experiments describing binding of 3H—ouabain to isolated myocytes in the

2

presence of Ca + at millimolar concentrations have not been previously reported.

Similar experiments examining 3H-ouabain binding to intact atrial muscle from

2+

guinea-pig heart in the presence of Ca and K+ (Herzig g a_1., 1985) or

ventricular muscle from guinea-pig heart (Kjeldsen gt g_l., 1985) in the absence of

2

Ca + and K+ have been published. The Scatchard plot describing binding to atrial

muscle is non-linear and the authors suggest this indicates positive cooperativity
in 3H—ouabain binding (Herzig e_t a_1., 1985a,b). The highest concentration of 3H-
ouabain used in these experiments with guinea-pig atrial muscle was 1.0 uM so
the presence of the very low affinity site seen in preparations of guinea-pig
ventricular myocytes in this study cannot be confirmed. Binding of 3H—ouabain to
guinea-pig ventricular muscle has been examined in a system using Mg2+ and
vanadate (V043') to support binding (Kjeldsen e_t g” 1985). The Scatchard plot
describing 3H-ouabain binding in this system was non-linear when binding occurred

3H—

in 3H-ouabain concentrations of less than 5.0 uM. Higher concentrations of
ouabain (10.0 and 20.0 uM) did not promote greater specific binding. This may
indicate that Mg2+ and V043- cannot support 3H—ouabain binding to the very low
affinity binding site. Alternatively, the Na,K-ATPase may have low affinity for
the glycoside if it has a low affinity for phosphate. This would result in a lower
fraction of the enzyme existing in a phosphorylated (binding) form, and therefore,
a low glycoside binding rate. Vanadate has a higher affinity for phosphate binding
sites on Na,K-ATPase than phosphate itself does (Wallick e_t g” 1979). Vanadate
would, therefore, increase the apparent affinity of the enzyme for ouabain. If

this occurs, then all specific 3H-ouabain binding sites may be saturated by a 5 uM

concentration of 3H—ouabain when the binding is supported by vanadate. Lack of

107

a very low affinity binding site in intact guinea-pig ventricular muscle (Kjeldsen
e_t a_1., 1985) may also mean that identification of this site in guinea-pig myocytes
is due to an artifact. Each experimental design described above used a
dissociation system to estimate non-specific 3H-ouabain binding. Authenticity of
the low-affinity site for 3H—ouabain binding to myocytes shall be discussed later
in this section.

Non-linear Scatchard plots describing 3

H-ouabain binding to high affinity
sites in guinea-pig cardiac muscle have been produced by three independent
investigators using three different preparations (Herzig e_t g” 1985b; Kjeldsen e_t
gl_., 1985; present results). Possible explanations of the curvature include:
multiple independent binding sites with different affinity; negative cooperativity
in 3H-ouabain binding; positive cooperativity in 3H-ouabain binding; a combination
of cooperativity and binding sites with different affinity.

Multiple independent binding sites with different kinetic parameters for 3H-
ouabain binding are to be expected for the Na,K-ATPase of guinea-pig cardiac
muscle (Godfraind e_t a_1., 1980), and Na,K-ATPase in general (Inagaki e_t a_1., 1974;
Choi and Akera, 1977; Wellsmith and Lindenmayer, 1980; Yoda and Yoda, 1986).
The presence of K+ is a key determinant of the relative abundance of each type of
binding site (Choi and Akera, 197 7) and can cause one conformation of the enzyme
to predominate over the other by its presence (Hansen, 1976) or its absence
(Godfraind gt a_1., 1980) as shown by the conditions under which binding produces
linear Scatchard plots. The different effect of K+ reported by these investigators
may reﬂect different sources of the enzyme or different ligand conditions to

3H-ouabain are believed

support binding. Different affinities of Na,K-ATPase for
to reﬂect binding of the glycoside to different conformations of the enzyme
which exist transiently as intermediates in the reaction sequence (Yoda and Yoda,

1986). The existence of two conformations of Na,K-ATPase, in intact guinea-pig

108

3

myocytes, which can bind H—ouabain was confirmed by showing the relative

abundance of specific binding sites for 3

H-oubain was changed by the presence or
absence of 10 mM K+.

Non-linear Scatchard plots describing binding to Na,K-ATPase of guinea-pig
myocytes cannot be explained solely on the basis of multiple binding sites whose
relative abundance is affected by K+ concentration. Another factor may be
redistribution of K+ subsequent to ouabain binding. This causes K+ in the
incubation media to increase in a manner dependent on ouabain concentration.

The effect of K+ to inhibit 3

H-ouabain binding to myocytes suggest there may be
some negative cooperativity in binding caused by such a redistribution. It is
impossible to know what a Scatchard plot, describing 3H-ouabain binding to the
Na,K—ATPase in myocytes, would look like if this occurred. If K+ was the only ion
redistributed, then an upwardly concave curve which would become linear after
complete redistribution of K+ had occurred, is expected on a Scatchard plot.
However, Na+ must also be redistributed in myocytes when K+ is redistributed.
Therefore, an increase in intracellular Na+ concentration will increase apparent
affinity of the Na,K-ATPase for 3H—ouabain at the same time that the increase in
extracellular K+ concentration decreases apparent affinity of the enzyme. The
shape of the curve on the Scatchard plot will be determined by the relative
amount of change in the Na,K-ATPase affinity for 3H-ouabain that redistribution
of each ion can cause. The increase in K+ concentration with full Sodium-pump
inhibition was only 12% of the initial K+ concentration under conditions of the
incubation for Scatchard analysis. This corresponds to an absolute change in K+
concentration of approximately 0.5 mM and is unlikely to significantly increase
the turnover rate of the enzyme to a conformation which does not bind 3H-

ouabain. The change in intracellular Na+ concentration is expected to be at least

a 10-fold increase, therefore, the effects of Na+ redistribution are expected to be

109

predominant. Also, the presence of monensin shifted the concentration response
curve for K+ release from myocytes incubated with ouabain toward the right,
indicating a desensitization to the action of ouabain. This is inconsistent with the
effect of monensin to increase the degree of curvature of the Scatchard plot if K+
redistribution and negative cooperativity contribute to the non-linearity. Fur—
thermore, redistribution of K+ in a manner dependent on the concentration of

2+

ouabain occurs in the absence as well as in the presence of Ca . Since Scatchard

3

plots describing H-ouabain binding to high affinity sites in the absence of Ca2+

are relatively linear, a redistribution of K+ is probably not contributing to the
non—linearity of Scatchard plots describing binding in the presence of Ca2+.

The action of monensin to shift concentration response curves for K+
redistribution by ouabain to the right is surprising and warrants further investiga-

tion. Monensin stimulates binding by 3

H-ouabain, therefore, the sodium-pump
should experience a greater degree of inhibition in the presence of monensin than
in its absence. Also, monensin is not entirely specific for Na+; therefore, some
K+ should be carried out of the cell by this ionophore. One feasible mechanism by
which monensin may cause the shift being discussed is to increase the driving
force for K+ accumulation by the Na,K-ATPase. An increase in Na,K-ATPase
turnover is also consistent with the increased curvature of the Scatchard plot if a
positive cooperativity caused by an increase in intracellular Na+ concentration is
responsible for the non-linearity.

The presence of two binding conformations of the Na,K-ATPase may also
contribute to the non-linearity of Scatchard plots describing 3H-ouabain binding

2+ the

to a high affinity site(s). Even when binding occurs in the absence of Ca
high affinity portion of the Scatchard plot is not really linear. This non-linearity
is furﬂier accentuated by the presence of monensin during binding. Monensin does

not have a significant effect to increase 3H-ouabain binding to cells incubated in

110

the absence of Ca2+ but it seems to change the shape of the Scatchard plot in the
middle portion of the curve. The interpretation that this is due to the presence of
two binding sites is highly speculative and may not be justified due to insensitivity

of this method of analysis for discerning small differences in receptor affinity.

3

Nevertheless, the subtle change in the shape of the Scatchard plot describing H-

2

ouabain binding to guinea-pig myocy tes caused by monensin when Ca + is absent

does support the possibility that Na,K-ATPase exists in two high affinity

conformations under these conditions.

3

In the presence of Ca2+, monensin stimulates binding of H-ouabain to

myocytes and increases the degree of curvature of the Scatchard plot describing
this binding. Monensin is presumed to do this by increasing Na+ inﬂux. This

mechanism of action is consistent with the effect of increasing Na+ inﬂux to

3

stimulate H—ouabain binding in intact cardiac muscle (Yamamoto gt 9, 1980;

3

Temma and Akera, 1982). Binding of H-ouabain to myocytes is not significantly

affected by monensin when binding occurs in the absence of Ca2+. This suggests

3

that each of these interventions increase binding of H-ouabain to the high

affinity site by the same mechanism, namely an increase in Na+ inﬂux. This

interpretation, however, is not consistent with the rightward shift in the ouabain

2+ removal (Figure 25). If the

2

concentration response curve caused by Ca

hypothesis that both the presence of monensin and the removal of Ca + stimulate

3H-ouabain binding by causing an increase in intracellular Na+ concentration is
correct, then either Ca2+ has a direct effect to decrease the reserve capacity of
the sodium-pump or the data indicating a rightward shift in the ouabain
concentration response curve as a result of Ca2+ removal are inaccurate. An
effect of Ca2+ to increase the permeability of the sarcolemma of cardiac

myocytes from rat heart to Na+ has been reported (Hohl e_t g, 1983), and is

consistent with data showing Na+ activity in sheep Purkinje fibers increases

111

approximately 50% when there is a 10-fold decrease in extracellular Ca2+

concentration over the range of Ca2+ concentrations from 0.2 to 16 mM (Ellis,
1977; Deitmer and Ellis, 1978). Elimination of the upwardly concave non-linearity
of Scatchard plots by increasing intracellular Na+ concentration is consistent with
positive cooperativity in cardiac glycoside binding to Na,K-ATPase in intact
cardiac muscle as proposed by Lullman and coworkers (Herzig e_t g1., 1985a,b).
The reason monensin does not promote linearity of Scatchard plots itself may be
because it does not raise the intracellular Na+ concentration significantly until

the reserve capacity of the sodium pump is eliminated. Removal of Ca2+

, on the
other hand may cause intracellular Na+ to increase even though the reserve
capacity of the sodium-pump is not eliminated. How this is possible is not
immediately obvious but this may indicate that Ca2+ has some effect to regulate
the activity of the sodium-pump. This possibility is supported by the data
indicating that removal of Ca2+ causes a rightward shift in the ouabain concen-
tration response curve for sodium-pump inhibition.

Positive cooperativity in binding generally is described by convex Scatchard
plots as shown in Figure 26. Binding of cardiac glycosides to Na,K-ATPase in
intact tissue is expected to exhibit posiﬁve cooperativity (Herzig gt Q” 1985a,b).
However, the cooperativity in binding will be by a different mechanism than
normally seen. As the sodium-pump reserve capacity is reduced by digitalis
binding, the turnover of the remaining Na,K-ATPase enzymes will increase. That
is, the number of times an individual enzyme will catalyze the transfer of Na+ and
K+ in any given unit of time will be increased. This means the amount of time the
individual enzyme exists in a binding configuration will increase. The number of

binding sites, therefore, effectively increases without any change in k1 or k'1 for

individual receptors. This is much different from the "normal" type of coopera-

tivity which results from partial occupancy of a population of receptors affecting

112

  
     
 

10
8
7 Increase Affinity
6 s
E
E’s 5 ’
Ll ' Increase
Sites
3
2
1
o

 

 

012 34 5678 910
Bound/Free

Figure 26. Scatchard plots describing binding to one population of receptors
without cooperativity in binding ---, or — positive cooperativity resulting from
increased affinity for the ligand, or with positive cooperativity resulting from
unmasking of binding site.

113

the affinity of the remaining unoccupied receptors for a ligand. Positive
cooperativity caused by an "unmasking" of receptor sites should be described by a
Scatchard plot in which the curve is below the normal straight line because the
low intracellular Na+ concentration reduces the amount of time the enzyme exists
in a binding conformation, and therefore, reduces the "net" binding rate. When
intracellular Na+ concentration has increased to an extent that all the receptors
are in a binding conformation, the plots should be superimposable.

2+ causes an unmasking of 3

If the absence of Ca H-ouabain binding sites in
this manner then the linear Scatchard plot describing the binding could be
considered an accurate description of the ouabain-enzyme interaction under
conditions of maximum time for enzyme existence in a binding conformation. The
unmasking may be an effect of increased intracellular Na+ concentration or some
direct effect of Ca2+ or a Ca2+-binding protein to alter affinity of the enzyme.
Binding in the presence of Ca2+ would then be a situation where the binding site
number is effectively increased by 3H—ouabain binding, resulting in positive
cooperativity. The effective increase in binding site number is not an increase in
total binding sites or pump units but merely indicates the enzymes which are
present exist for a greater amount of time in a Na+-induced binding conformation
as more ouabain is specifically bound. This is because there is an increase in
intracellular Na+ concentration caused by 3H-ouabain binding. Increased curva-
ture of Scatchard plots when binding occurs in the presence of Ca2+ and monensin
indicates that monensin simply amplifies the unmasking effect of ouabain. With

this amplification the curves describing 3

H-ouabain binding to myocy tes in a
system where there is positive cooperativity and one where there is no cooperati-
vity Will be superimposed at a lower 3H—ouabain concentration. Scatchard plots
describing 3H-ouabain binding to high affinity sites on guinea-pig myocytes are in

good agreement with this model of positive cooperativity in the binding of the

114

cardiac glycosides. Deviation of the experimental data from the model is not
serious, considering the error involved in estimating non-specific binding and the
potential contribution of multiple binding sites and changes in conformation which

can affect 3

H-ouabain binding.

Authenticity of the low affinity binding site for 3H—ouabain is questionable.
This site was not shown to be saturable and may represent an artifact of the
method used to estimate non-specific binding. At concentrations of 3H-ouabain in
excess of 3.0 uM, non-specific binding was greater than 20% of the specific
binding and at the highest concentration of 3H-ouabain used (20 uM) non-specific
binding was estimated to account for 37% of total binding when binding occurred
in the absence of Ca2+.

Accuracy of estimates for non-specific binding is a serious concern during
interpretation of data describing 3H-ouabain binding to myocytes incubated in
high concentrations of 3H-ouabain. As discussed earlier, non-specific binding can
be underestimated when the amount of 3H-ouabain remaining bound after 60 min

3

of dissociation is used as an estimate of non-specific binding. Binding of H-

ouabain to a low affinity site in intact, guinea-pig ventricular muscle was not
detected when binding was supported by Mg2+ and V043'. It is possible these
ligands cannot support binding to low affinity forms of the Na,K-ATPase, or that
vanadate increases the affinity of low affinity binding sites. Alternatively, the
low affinity binding site detected in guinea-pig myocytes may be artifactual.
Support for the authenticity of the low affinity binding site for 3H-ouabain
is the increase in 3H—ouabain bound when myocytes are incubated in the absence

3

of K+. Total binding to myocytes following a 40-min incubation with 2.5 uM H—

ouabain in the absence of K+ was 23.2:0.7 pmol/mg protein. After 30 sec of
dissociation at 37°C these myocytes contained 13.4:0.7 pmol/mg protein of 3H—

ouabain. This rapid dissociation may represent release of 3H-ouabain bound to the

115

low affinity site described in Scatchard plots. Additional experiments would be
necessary to substantiate this point.
Even though the data support the hypothesis that there is positive coopera-

tivity in 3H-ouabain binding to guinea-pig myocy tes in the presence of Ca2+ which

is caused by changes in intracellular Na+ concentration, it is possible that Ca2+

3

has a direct effect on H-ouabain binding which causes Scatchard plots to be non-

2

linear. In preparations of isolated Na,K-ATPase, Ca + inhibits enzyme activity by

competition with Na+ (Tobin gt g” 1973). This inhibition, however, should not be

significant at intracellular Ca2+ concentrations present in quiescent cardiac

2

muscle because the IC50 for Ca + inhibition of Na,K-ATPase activity is approxi-

mately 0.5 mM (Tobin e_t gl_., 1973; Godfraind gt gt, 1977; Beauge and Campos,

1983; Powis e_t $1., 1933). The possibility that Ca2+ acts through a Ca2+-binding

protein to affect the affinity of Na,K-ATPase for ouabain has been discussed in

the Introduction section of this work.

2

Direct actions of intracellular Ca + apparently do not increase or decrease

3

affinity of Na,K-ATPase for H-ouabain in guinea-pig myocytes. This is indicated

by the fact that the Ca2+ ionophore, A23187, does not affect 3H—ouabain binding

2

when cells are incubated in solution containing 1.8 mM Ca + and 3.8 mM K+.

Under these conditions, intracellular Ca2+ concentration is low as judged by

2

myocytes being quiescent. A2318? should increase intracellular Ca + concentra-

tion under these conditions. Furthermore, when cells are incubated in solution

2+

containing 1 mM K+ and 0.1 mM Ca , the myocytes contract spontaneously,

indicating that the intracellular Ca2+ concentration is increased. Yet under these

incubation conditions, 3H-ouabain binding is stimulated. It is possible that binding

2

of 3H—ouabain in solutions containing 1 mM K+ and 0.1 mM Ca + is increased

because of the effect of the low K+ concentration to promote the enzyme to exist

in a Na+-induced binding conformation, and that Ca2+ actually is inhibiting

116

binding. This possibility was not tested directly, but if this was the case, 3H—

ouabain binding should have been stimulated by lanthanum, which blocks Ca2+

entry. Because lanthanum had no effect on 3H—ouabain binding, the available

2+ does not have an effect on cardiac glycoside binding.

2

evidence suggests that Ca

2

Actions of Ca +-dependent protein kinases or other Ca +-binding proteins

also are without effect to increase or decrease affinity of the Na,K-ATPase for

3 2+ concentration is

H—ouabain. This is deduced because when intracellular Ca
elevated by incubating myocytes in solution with 1 mM K+, there are no effects
on 3H-ouabain binding by agents which increase (A2318?) or decrease (lanthanum)
intracellular Ca2+ concentration, or by agents which promote phosphorylation of
sarcolemmal constituents (isoproterenol, TPA).

Because no direct action of Ca2+ or a Ca2+—dependent protein kinase can be

detected, the effect of EGTA to stimulate 3

2+

H-ouabain binding must be attributed

to an indirect effect of Ca removal.

3

Data from experiments examining H-ouabain binding to guinea-pig myo-

cytes indicate that there are two or three classes of binding sites for cardiac
glycosides corresponding to different conformations of the Na,K-ATPase. The
total number of high affinity 3H-ouabain binding sites in guinea-pig cardiac

myocytes was estimated to be approximately 14 pmol/mg protein. This estimate

was made from analysis of the Scatchard plot describing 3

the myocytes incubated in the absence of Ca2+. Fourteen pmol/mg protein

corresponds to approximately 40 million 3H—ouabain binding sites per cell. These

H-oaubain binding to

sites have an affinity (KD) for 3H—ouabain of 0.4-0.7 uM in the absence of Ca2+

and a range of apparent affinity from 0.15-1.9 uM in the presence of Ca2+.

Estimates of affinity (KD) for 3H—ouabain were made by analysis of Scatchard

plots. Analysis of the curve describing 3H—ouabain binding in the presence of

2

Ca + was done by drawing tangents to the curve where it was most linear (KD

117

estimate of 0.15 uM) and prior to the linear portion of the curve which describes
binding to the very low affinity site (KD estimate of 1.9 uM). There is good

evidence for positive coopera tivity in 3H—ouabain binding which is observed only

2+. Direct effects of Ca2+, either by

3

when binding occurs in the presence of Ca

2

the ion itself or via a Ca +-binding protein, do not appear to affect H—ouabain

binding to guinea-pig myocy tes.

D. Estimation of Sodium-Pump Activity and Capacity in Guinea-pig Cardiac
Myogy tes

Measuring ouabain-sensitive

 

86 42

Rb+ or K+ uptake is a well established
method for estimating sodium—pump activity (Akera and Brody, 1985). Experi-
ments with intact cardiac muscle have established that unless Na+ is accumulat-
ing in the tissue, or has accumulated to the extent that Na+ is not the rate-
limiting substrate for Na,K-ATPase activaﬁon, then sodium-pump activity is
determined by Na+ inﬂux rate or intracellular Na+ concentration (Yamamoto e_t
g” 1979; Akera gt 9, 1981). If sodium-pump activity is determined by substrate
availability, it will be impossible to detect a direct effect of any intervention on
the Na,K-ATPase by monitoring sodium-pump activity. Therefore, interventions
which may affect the time course of the sodium transient, an early event in
contraction of cardiac muscle (Akera and Brody, 1978), by affecting performance
of the Na,K-ATPase during the Na+ transient, a time at which enzyme activity is
not limited by Na+ concentration, cannot be detected by measuring effects on
sodium-pump activity but only by measuring effects on sodium—pump capacity
(Akera and Brody, 1985).

Sodium-pump activity in intact cardiac muscle is stimulated by interven-
tions which increase Na+ inﬂux (Yamamoto gt Q” 1979; Yamamoto _e_t g_l., 1980;

Akera e_t a_1., 1981). Theoretically, increasing Na+ inﬂux to a large degree, will

eliminate the reserve capacity of the sodium-pump so that the capacity of the

118

pump can be estimated (Akera and Brody, 1985). This has been previously
attempted with intact muscle preparations and shifts in the ouabain concentration
response curve were detected (Yamamoto _e_t a_1., 1980).

Direct comparison of the magnitude of 86Rb+ ﬂuxes between myocytes and
intact tissue would require estimates of total cell surface area for each
preparation and estimates of the relative abundance of muscle to non-muscle cells
as well as the ﬂux into each type of cell for the intact tissue. These estimates
are not available, therefore, only relative changes in ion ﬂuxes caused by similar
interventions can be compared. The most effective means for increasing ouabain-

86

sensitive Rb+ uptake into intact muscle preparations is to electrically stimulate

those preparations (Yamamoto _e_t a_1., 1979; Akera gt g” 1981). The Na+

86Rb+ uptake

ionophore, monensin, also causes an increase in ouabain-sensitive
into intact muscle preparations, but the magnitude of the increase is only about
25% of that achieved with electrical stimulation (Yamamoto gt g, 1979). Thus,

86Rb+ uptake into intact tissue which has

the largest increase in ouabain-sensitive
been reported, is caused by electrical stimulation and corresponds to an increase
in sodium-pump activity of approximately 200% above that which is observed in
quiescent tissue (Yamamoto e_t g” 1979; Akera e_t a_1., 1981).

86Rb+ uptake, relative to the rate

Stimulated increases in ouabain-sensitive
observed in quiescent preparations of each type, are much larger in myocytes than
the values reported for intact tissue. Technical problems with electrical
stimulation of myocytes precluded use of this method to enhance Na+ inﬂux and
therefore, sodium-pump activity. However, the maximum relative increase with
monensin or with sodium-loading was approximately 400% or twice that which was
achieved with electrical stimulation of intact tissue. Similar attempts to

maximally stimulate sodium-pump activity in cultured embryonic cells or disso-

ciated myocy tes from hearts of mature animals have not been reported.

119

86Rb+ uptake

The effectiveness of monensin to stimulate ouabain-sensitive
in guinea-pig myocytes is distinctly different from its modest ability to do so in
intact muscle and the limited effective concentration range of monensin (< 2 uM)
in intact muscle preparations (Yamamoto _e_t gt” 1979). This limited effective
concentration range may be due to the longer incubation and up take period used in
intact muscle (generally 30 min) which usually allow toxic effects of high
concentrations of monensin to be seen. The relatively modest effect of monensin
to stimulate sodium-pump activity (maximum increase is 50%) in intact tissues
may be due to diffusion barriers which limit ion ﬂuxes. It is unlikely that the
isolated myocy tes have a lower basal Na+ inﬂux rate than that which exists for
the cells in intact tissue (a factor which would cause the relative increase to be
larger than the absolute increase). Therefore, the ability to stimulate ouabain-

86

sensitive Rb+ uptake in myocytes beyond what has been achieved with intact

tissue probably represents a real increase in sodium-pump activity. The most

likely cause for this is the elimination of diffusion barriers which can limit the

86

movement of Rb+ in intact tissue.

86Rb+ uptake into myocytes ultimately

The plateau in ouabain—sensitive
achieved at high concentrations of monensin or with long periods of sodium-
loading suggests that intracellular Na+ has increased to the degree that it is no
longer rate-limiting for sodium—pump activity. The fact that each of these
interventions produce a maximum effect of approximately 400% over basal values
to increase sodium-pump activity, also suggests that reserve capacity has been
eliminated by each intervention. This must be established by an independent
method, however, such as measuring an increase in cellular Na+ load without a

86

concomitant increase in ouabain-sensitive Rb+ uptake.

86

The effect of Ca2+ to inhibit ouabain—sensitive Rb+ uptake has not been

previously reported. This effect of Ca2+ was significant in quiescent myocytes,

120

myocy tes subjected to all degrees of sodium-loading, and cells exposed to low or
moderate concentrations of monensin; but not for myocytes exposed to high
concentrations of monensin. Possible mechanisms for this effect of Ca2+ to
inhibit sodium-pump activity are: a direct action on the Na,K—ATPase; competi-
tion for intracellular Na+ via the Na/Ca exchange mechanism, or altered
membrane permeability to Na+ (Hohl gt gl_., 1983). Because the ouabain-sensitive
86Rb+ uptake is an estimate of Na+ inﬂux unless the sodium—pump is operating at
capacity (Akera gt gl., 1981; Akera and Brody, 1985), the effect of Ca2+ to
decrease sodium-pump activity in quiescent myocytes and those subjected to
moderate concentrations of monensin or short periods of Na+-loading must be due
to either an altered sarcolemmal permeability or competition between the Na,K—
ATPase and the Na/Ca exchanger for intracellular Na+. Sarcolemmal permeabi-
lity to Na+ should not affect sodium-pump activity when the myocytes are

2+ to inhibit sodium-pump

maximally sodium-loaded. Therefore, the effect of Ca
activity in sodium-loaded myocytes is inconsistent with this explanation for the
action of Ca2+. If, however, the intracellular Na+ concentration continues to
increase with time as myocy tes are sodium-loaded, then competition for intracel-
lular Na+ via the Na/Ca exchange cannot explain why the plateau in ouabain-

86Rb+ uptake occurs at the same time for sodium-loading but at a lower

86

sensitive
value for activity of the sodium—pump when the Rb+ uptake is monitored in
solution containing Ca2+. Thus, each explanation for the effect of Ca2+ on
sodium-pump activity is inconsistent with some aspect of the data. It is possible
that each of the three mechanisms is functioning under some of the conditions
examined. Further experiments are required in order to draw firm conclusions

2 86+

. . . + . . .
concerning the mechanism by which Ca decreases ouabain-senSitive Rb

uptake.

121

Concentra tion-dependent inhibition of sodium-pump activity by ouabain has
been demonstrated in intact tissues (Yamamoto e_t a_1., 1980), isolated myocytes
(Silver and Houser, 1985) and cultured cardiac cells (McCall, 1979; Werden e_t a_1.,
1983; Kazazoglou e_t a_1., 1983). It has also been demonstrated, in intact cardiac
muscle, that the ouabain concentration response curve is shifted to the left by
interventions which increase Na+ inﬂux and sodium—pump activity (Yamamoto gt
a_1., 1980). These data are in good agreement with the shift to the left caused by
monensin of the ouabain concentration response curves for inhibition of sodium-
pump activity. These data from intact tissue and myocytes have been interpreted
to indicate that augmented Na+ inﬂux decreases the reserve capacity of the
sodium-pump. As pointed out by Yamamoto and coworkers (1980), however, an
increase in ouabain binding caused by the increased Na+ inﬂux (Temma and Akera,
1982; Kennedy e_t a_1., 1983) can be at least partially responsible for the shift to

the left of this curve.

SUMMARY AND CONCLUSIONS

Preparations of Ca2+—tolerant myocytes can be isolated from hearts of adult
animals after digesting connective tissue in the hearts with collagenase and
hyaluronidase. These preparations are composed of a mixed population of viable
and non-viable myocytes from which the living myocytes can be selected to
greater than 80% purity by elutriation and gravity sedimentation.

Binding of 3H-ouabain for the estimation of kinetic parameters of binding
can be examined using these myocytes. Accurate estimation of non-specific
binding to myocytes requires that a dissociation method be used if bindirg occurs
in the presence of Ca2+ and K+. This is in contrast to experiments utilizing
isolated enzyme or tissue homogenate preparations where non-specific binding can

be estimated using an excess of non-labeled drug.

3 2+

H-ouabain binding to myocytes in the presence of Ca and K+ at

millimolar concentrations is described by non-linear Scatchard plots. This

indicates that there is more than one binding site for 3H-ouabain or that there is

3

cooperativity in binding. Each of these aspects of H-ouabain binding preclude

3

use of the displacement method for estimating kinetic parameters at H—ouabain

binding. In the absence of Ca3+, 3

H-ouabain binds to guinea-pig myocytes in such
a manner that two distinct linear components of binding can be discerned on
Scatchard plots; a high affinity site with a KD of 0.5 uM and a Bmax equal to 14.0
pmol/mg protein and a low affinity site with a KD of 3.4 11M and a Bmax equal to
8.4 pmol/mg protein. The high affinity site probably represents the Na,K-ATPase

as evidenced by the ability of K+, when present during the binding reaction at a

122

123

3

concentration of 10 mM, to decrease incorporation of H-ouabain into this site.

The nature of the low affinity site is uncertain and it is possible that this

represents an artifact of the method used to estimate non—specific binding.

3

Non-linearity of Scatchard plots describing H-ouabain binding to the high

affinity site in guinea-pig myocytes when binding occurs in the presence of Ca2+

and K+, is probably caused by positive cooperativity in binding under these

incubation conditions. The cooperativity results from an inactivation of sodium-

. O O 0 O O O +
3H—ouaba1n binding. This causes an increase in intracellular Na

3

pump units by
concentration which is a key determinant promoting H—ouabain binding. Positive
cooperativity by this mechanism is described by Scatchard plots which are
upwardly concave. The possibilities that multiple binding sites or negative
cooperativity associated with K+ redistribution were contributing to the non-
linearity of the Scatchard plots describing 3H-ouabain binding to myocytes were
examined. Each of these phenomena may contribute toward the non-linearity,
however, neither can have a large inﬂuence because when binding occurs in the
absence of Ca2+, Scatchard plots are linear but these phenomena still occur. The

3+ itself or a Ca3+-dependent process inhibited

2+

possibility that intracellular Ca
3H-ouabain binding was examined by determining if agents which increase Ca
inﬂux (A2318?) decrease Ca3+ inﬂux (lanthanum) or stimulate a protein kinase
(isoproterenol and TPA) affect 3H—ouabain binding. None of these agents had a

3 2+

significant effect on H-ouabain binding, therefore, a direct effect of Ca or a

Ca2

+-dependent process on the Na,K-ATPase to affect cardiac glycoside binding
appears to be unlikely.
Sodium-pump activity can be readily measured in preparations of myocytes

using ouabain-sensitive 86Rb+ uptake as an index of pump activity. Sodium-pump

activity is stimulated by the presence of monensin in a concentration—dependent

manner with maximum uptake exceeding basal uptake by 400%. This degree of

124

stimulation is also observed following incubation of myocytes in a K+-free Rb+-

2

free solution containing 10 uM Ca + for 45 min or longer. The high ouabain-

86

sensitive Rb+ uptake seen when myocytes are exposed to high concentrations of

monensin or long periods of sodium-loading may represent the capacity of the
sodium-pump in intact cells. The shift to the left of the concentration response
curve for sodium—pump inhibition by ouabain which is caused by monensin

represents a decrease in the reserve capacity of the sodium-pump.

2+

The presence of Ca in incubation media during determination of ouabain-

86

sensitive Rb+ uptake inhibits both basal and stimulated uptake. The mechanism

for this inhibition is unknown.

BIBLIOGRAPHY

BIBLIOGRAPHY

Adams, R.J., Schwartz, A., Grupp, G., Grupp, 1., Lee, S—W. and Wallick, E.T.:
High-affinity ouabain binding site and low-dose positive inotropic effect in
rat myocardium. Nature 2%: 167-169, 1982.

Aiton, J.F., Lamb, J.F. and Ogden, P.: Down-regulation of the sodium pump
following chronic exposure of HeLa cells and chick embryo heart cells to
ouabain. Br. J. Pharmacol. Q: 333-340, 1981.

Akera, T. and Brody, T.M.: Membrane adenosine triphosphatase: The effect of
potassium on the formation and dissociation of the ouabain-enzyme com-

Akera, T. and Brody, T.M.: The role of Na+,K+-ATPase in the inotropic action of
digitalis. Pharmacol. Rev. E: 187—220, 1978.

Akera, T. and Brody, T.M.: Myocardial membranes: Regulation and function of
the sodium pump. Ann. Rev. Physiol. gt: 375-388, 1982.

Akera, T. and Brody, T.M.: Estimating sodium pump activity in beating heart
muscle. Trends Pharmacol. Sci. 6: 156-159, 1985.

Akera, T. and Cheng, V-J.K.: A simple method for the determination of affinity
and binding site concentration in receptor binding studies. Biochim.
Biophys. Acta 4_70_: 412-423, 1977.

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. ﬂ: 101-114, 1976.

Akera, T., Tobin, T., Gatti, A., Shieh, I-S. and Brody, T.M.: Effect of potassium
on the conformational state of the complex of ouabain with sodium- and
potassium-dependent adenosine triphosphatase. Mol. Pharmacol. Q: 509-
518, 1974.

Akera, To, YamamOtO, So, Temma, K0, Kim, D-Ho and BPOdy, T.M.: IS Ouabain-

sensitive rubidium or potassium uptake a measure of sodium pump acivity in
isolated cardiac muscle? Biochim. Biophys. Acta ﬁg: 779-? 90, 1981.

Amer, B. .: The receptor function of the Na+iK+’iIGtiVated adenosine t1‘iPhOS"'
phatase system. Biochem. J. 221: 1-11, 1985.

1‘25

126

Baker, J.F., Bullock, G.R. and Hearse, D.J.: The temperature dependence of the
calcium paradox: Enzymatic, functional and morphological correlates of
cellular injury. J. Mol. Cell. Cardiol. _1_5: 393-411, 1983.

Beauge, L. and Camp , MtA.: Calcium inhibition of the ATPase and phosphatase
activities of (Na +K )-ATPase. Biochim. Biophys. Acta m: 137-149, 1983.

Bentfeld, M., Lullmann, H., Peters, T. and Proppe, D.: Interdependence of ion
transport and the action of ouabain in heart muscle. Br. J. Pharmacol. 6_1:
19-27, 1977.

Berridge, M.J.: Inositol triphosphate and diacylglycerol as second messengers.
Biochem. J. Qt); 345-360, 1984.

Berridge, M.J. and Irvine, R.F.: Inositol triphosphate, a novel second messenger
in cellular signal transduction. Nature it}: 315-321, 1984.

Bihler, 1., Ho, T.K. and Sawh, P.C.: Isolation of Cam-tolerant myocytes from
adult rat heart. Can. J. Physiol. Pharmacol. Q: 581-588, 1984.

Bkaily, G., Sperelakis, N. and Doane, J.: A new method for preparation of
isolated single adult myocytes. Am. J. Physiol. 221: H1018-H1026, 1984.

Blayney, L., Thomas, H., Muir, J. and Henderson, A.: Action of caffeine on
calcium transport by isolated fractions of myofibrils, mitchondria, and
sarcoplasmic reticulum from rabbit heart. Circ. Res. 43: 520-526, 1978.

Bluschke, V., Bonn, R. and Greeff, K.: Increase in the (Na++K+)-ATPase activity
in heart muscle after chronic treatment with digitoxin or potassium
deficient diet. Eur. J. Pharmacol. 3_?: 189-191, 1976.

Bradford, M.M.: A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal.
Biochem. Q: 248-254, 1976.

Bruckner, R., Meyer, W., Mugge, A., Schmitz, W. and Scholz, H.: a-Adrenocep—
tor-mediated positive inotropic effect of phenylephrine in isolated human
ventricular myocardium. Eur. J. Pharmacol. ﬂ: 345-347, 1984.

Burgisser, E.: Radioligand-receptor binding studies: What‘s wrong with the
Scatchard analysis? Trends Pharmacol. Sci. g: 142-144, 1984.

Busse, F., Lullman, H. and Peters, T.: Concentration dependence of the binding
of ouabain to isolated guinea pig atria. J. Cardiovasc. Pharmacol. _1_: 670-
698, 1979.

Charlemagne, D., Leger, J., Schwartz, K., Geny, B. Zachowski, A. agid Lelievre,
L.: Involvement of tropomyosin in the sensitivity of Na +K -ATPase to
ouabain. Biochem. Pharmacol. 2_9: 297-300, 1980.

Choi, Y.Rg and Akera, T.: Kinetic studies on the interaction between ouabain and
(Na ,K )-ATPase. Biochim. Biophys. Acta gt: 648-659, 1977. '

127

Cohen, 1., Falk, R. and Gintant, G.: Saturation of the internal sodium site of the
sodium pump can distort estimates of potassium affinity. Biophys. J. at:
719-727, 1984.

Cook, J.S., Tate, E.H. and Shaffer, C.: Uptake of [3Hlouabain from the cell
surface into the lysosomal compartment of HeLa cells. J. Cell. Physiol.
tttt: 84-92, 1982.

Daut, J. and Rudel, R.: The electrogenic sodium pump in guinea-pig ventricular
muscle: Inhibition of pump current by cardiac glycosides. J. Physiol. 310:
243-264, 1982.

Dietmer, 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. 2E: 437-453, 1978.

Dutta, S. and Marks, B.H.: Factors that regulate ouabain-H3 accumulation by the
isolated guinea-pig heart. J. Pharmacol. Exp. Ther. ﬂ: 318-325, 1969.

Eisner, D.A. and Lederer, W.J.: The role of the sodium pump in the effects of
potassium—depleted solutions on mammalian cardiac muscle. J. Physiol. M:
279-301, 1979.

Eisner, D.A. and Lederer, W.J.: Characterization of the electrogenic sodium
pump in cardiac purkinje fibres. J. Physiol. 3_0_3: 441-474, 1980.

Eisner, D.A. and Valdeolmillos, M.: The mechanism of the increase of tonic
tension produced by caffeine in sheep cardiac purkinje fibers. J. Physiol.
3_65: 313-326, 1985.

Ellis, D.: the effects of external cations and ouabain on the intracellular sodium
activity of sheep heart purkinje fibers. J. Physiol. m: 211-240, 197?.

Erdmann, E., Bolte, H.D. and Luderitz, B.: The (Na++K+)-ATPase activity of
guinea pig heart muscle in potassium deficiency. Arch. Biochem. Biophys.
t4_5_: 121-125, 1971.

Erdmann, E., Patzelt, R. and Schoner, W.: The cardiac glycoside receptor: Its
properties and it§ cgrrelation to nucleotide binding sites, phosphointer-
mediate, and (Na +K )-ATPase activity. In Recent Advances in Studies (a
Cardiac Structure and Metabolism. The Sarcolemma7P-E. Roy and N.S.
Dhalla, eds.), Vol. 9, pp. 329-335. Baltimore: University Park Press, 1976.

 

 

Erdmann, E., Philipp, G. and Scholz, H.: Cardiac glycoside receptor, (Na++K+)-
ATPase activity and force of contraction in rat heart. Biochem. Pharmacol.
2_9_: 3219-3229, 1980.

Fabiato, A. and Fabiato, F.: Contractions induced by a calcium-triggered release
of calcium from the sarcoplasmic reticulum of single skinned cardiac cells.
J. Physiol. (Lond.) 2%: 25-43, 1970.

Farmer, B.B., Harris, R.A., Jolly, W.W., Hathaway, D.R., Katzberg, A., Watanabe,
A.M., Whitlow, A.L. and Besch, H.R. Jr.: Isolation an characterization of
adult rat heart cells. Arch. Biochem. Biophys. 1_?9: 545-558, 1977.

128

Farmer, B.B., Mancina, M., Williams, E.S. and Watanabe, A.M.: Isolation of
calcium tolerant myocytes from adult rat hearts: Review of the literature
and description of a method. Life Sci. gt: 1-18, 1983.

Fricke, U. and Klaus, W.: Eviden e fpr two different Na+-dependent [3H]-
ouabain binding sites of a Na -K -ATPase of guinea-pig hearts. Br. J.
Pharmacol. gt: 423-428, 1977.

Friedman, 1., Schwalb, H., Hallaq, H., Pinson, A. and Heller, M.: Interactions of
cardiac glycosides with cultured cardiac cells. Biochim. Biophys. Acta 598:
272-284, 1980.

Gadsby, D.C. and Cranefield, P.F.: Direct measurement of changes in sodium
pump current in canine cardiac purkinje fibers. Proc. Natl. Acad. Sci. USA
1g: 1783-1787, 1979.

Gelbart, A. and Goldman, R .: Correlation between microsomal (Na++K+)-
ATPase activity and [ H]ouabain binding to heart tissue homogenates.
Biochim. Biophys. Acta 581: 689-694, 1977.

Geny, B., Peraf, A., Fedon, Y. and Charlemagne, .: Characterization of a B-
actinin-like protein in purified non-muscle cell membranes. Biochim.
Biophys. Acta 912.: 345-354, 1982.

Gervais, A., Lane, L.K., Anner, B.M., Lindenmayer, G.E. and Schwartz, A.: A
possible molecular mechanism of the action of digitalis. Circ. Res. 39: 8-14,
197?.

Ghysel-Burton, J. and Godfraind, T.: Stimulation and inhibition of the sodium
pump by cardioactive steroids in relation to their binding sites and their
inotropic effect on guinea-pig isolated atria. Br. J. Pharmacol. g9: 175-184,
1979.

Godfraind, T., Depover, A. and Lutete, D-N.T.: Identification with potassiitgn apd
vanadate of two classes of specific ouabain binding sites in a (Na +K )-

ATPase preparation from the guinea-pig heart. Biochem. Pharmacol. 2_9_:
1195-1199, 1980.

Godfraind, T., Depover, A. and Verbeke, .: Inﬂuence of pH and sodium on the
inhibition of guinea-pig heart (Na +K )-ATPase by calcium. Biochim.
Biophys. Acta 581: 202-211, 1977.

Godfraind, T. and Ghysel-Bruton, J.: Binding sites related to ouabain-induced
stmulation or inhibition of the sodium pump. Nature 26_5_: 165-166, 197?.

Godfraind, T. and Ghysel-Burton, J.: Independence of+ thg positive inotropic
effect of ouabain from the inhibition of the heart Na /K pump. Proc. Natl.
Acad. Sci. USA 11: 3067-3069, 1980.

Godfraind, T., Ghysel-Burton, J. and Depover, A.: Dihydroouabain is an
antagonist of ouabain inotropic action. Nature 2%: 824-826, 1982.

Godfraind, T. and Lesne, M.: The uptake of cardiac glycosides in relation to their
actions in isolated cardiac muscle. Br. J. Pharmacol. 4_6: 488-497, 1972.

129

Hansen, +0.: + Non-uniform populations of g—strophanthin binding sites of
(Na +K )-activated ATPase. Biochim. Biophys. Acta 4_3§:383-392,1976.

Haworth, R.A., Hunter, D.R. and Berkoff, H.A.: The isolation of Ca3+-resistant
myocytes from the adult rat. J. Mol. Cell. Cardiol. E: 715-723, 1980.

Haworth, R.A., Hunter, D.R. and Berkoff, H.A.: Mergignism of Ca2+ resistance
in adult heart cells isolated with trypsin plus Ca . J. Mol. Cell. Cardiol.
15.: 523-530, 1982.

Herzig, S., Krey, U., Lullmann, H. and Mohr, K.: Is digitalis binding in intact
myocardium co-operative? Trends Pharmacol. Sci. _5_: 432-433, 1985a.

Herzig, S., Lullmann, H. and Mohr, K.: On the cooperativity of ouabain-binding
to intact myocardium. J. Mol. Cell. Cardiol. _1_1: 1095-1104, 1985b.

Hohl, C.M., Altschuld, R.A. and Brierley, G.P.: Effects of calcium on the
permeability of isolated adult rat heart cells to sodium. Arch. Biochem.
Biophys. a: 197-205, 1983.

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. 249: 5135-5140, 1974.

Isenberg, G.: Contractility of isolated bovine ventricular myocytes is enhanced
by intracellular injection of cardioactive glycosides. In Cardiac Glycoside
Receptors and Positive Inotropy (E. Erdmann, ed.), Steinkopff, Darmstadt,
pp. 56-71, 1984.

 

Isenberg, G. and Klockner, U.: Calcium tolerant ventricular myocytes prepared
by preincubation in a "KB medium". Pﬂugers Arch. 3g: 6-18, 1982.

Kazazoglou, T., Renaud, J-F., Rossi, B. and Lazdunski, M.: Two classes of
ouabairt receptors in chick ventricular cardiaf; cells and their r tion to
(Na ,K )-ATPase inhibition, intracellular Na accumulation, Ca influx,
and cardiotonic effects. J. Biol. Chem. ﬂ: 12163-12170, 1983.

Kennedy, R.H., Akera, T. and Brody, T.M.: How increased sodium inﬂux
enhances digoxin-induced arrhythmias in guinea-pig atrial muscle. Eur. J.
Pharmacol. Q: 199-209, 1983.

Kim, D., Marsh, J.D., Barry, W.H. and Smith, T.W.: Effects of growth in low
potassium medium or ouabain on membrane Na,K-ATPase, action transport,
and contractility in cultured chick heart cells. Circ. Res. §_5_: 39-48, 1984.

Kjeldsen, K., Norgaard, A., Hansen, (3 and Clausen, T.: Significance of skeletal
muscle digitalis receptors for [ H]ouabain distribution in the guinea pig. J.
Pharmacol. Exp. Ther. Q43 720-727, 1985.

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. _28_5: 185-200, 1974.

130

Ku, D.D., Akera, T., Tobin, T. and Brody, T.M.: Comparative spegieg studies on
the effect of monovalent cations and ouabain on cardiac Na ,K -adenosine
triphosphatase and contractile force. J. Pharmacol. Exp. Ther. 122: 458-
469, 1976.

Kuo, J.F., Andersson, R.G.G., Wise, B.C., Mackerlova, L., Salomonsson, 1.,
Brackett, N.L., Katoh, N., Shoji, M. and Wrenn, R.W.: Calcium-dependent
protein kinase: Widespread occurrence in various tissues and phyla of the
animal kingdom and comparison of effects of phospholipid, calmodulin, and
triﬂuoperazine. Proc. Natl. Acad. Sci. USA ﬂ: 7039-7043, 1980.

Langer, G.A.: Sodium-calcium exchange in the heart. Ann. Rev. Physiol. ﬂ:
435-449, 1982.

Lee, C.O., Abete, P., Pecker, M., Sonn, J.K. and Vassalle, M.: Strophanthidin
inotropy: Role of intracellular sodium ion activity and sodium-calcium
exchange. J. Mol. Cell. Cardiol. ﬂ: 1043-1053, 1985.

Lee, C.O. and Dagostino, M.: Effect of strophanthidin on intracellular Na ion
activity and twitch tension of constantly driven canine cardiac purkinje
fibers. Biophys. J. go: 185-198, 1982.

Lee, C.O., Kang, D.H., Sokol, J.H. and Lee, K.S.: Relation between intracellular
Na ion ach‘vity and tension of sheep cardaic purkinje fibers exposed to
dihydro-ouabain. Biophys. J. 29: 315-330, 1980.

Lee, K.S. and Klaus, W.: The subcellular basis for the mechanism of inotropic
action of cardiac glycosides. Pharmacol. Rev. 22: 193-242, 1971.

Lelievre, L.G., Potter, J.D., Piasick, M., Wallick, E.T., Schwartz, A., Charle-
magne, D. and Geny, B.: Specific involvement of calmodulin and +nog-
specific effect of tropomyosin in the sensitivity to ouabain of Na ,K -
ATPase in murine plasmocytoma cells. Eur. J. Biochem. 145: 13-19, 1985.

Lelievre, L., 4.Zac owski, A., Charlemagne, D., Laget, P. and Paraf, A.: Inhibition
of (Na +K )-ATPase by ouabain: Involvement of calcium and membrane
proteins. Biochim. Biophys. Acta 5_5_7_: 399-408, 1979.

Limas, C.J.: Characterization of phorbol diester binding to isolated cardiac
myocytes. Arch. Biochem. Biophys. £83 300-304, 1985.

Ling, E. and Sapirstein, V.: Phorbol ester stimulates the phosphorylation of
rabbit erythrocyte band 4.1. Biochem. Biophys. Res. Commun. 120: 291-298,
1984.

Ling, L. and Cantley, L.: The (Na,K)-ATPase of friend erythroleukemia cells is
phosphorylated near the ATP hydrolysis by an endogenous membrane-bound
kinase. J. Biol. Chem. 2_52: 4089-4095, 1984.

Lullman, H. and Peters, T.: Action of cardiac glycosides on the excitation-
contraction coupling in heart muscle. Prog. Pharmacol. 2: 5-57, 197 9.

131

Mansier, P., Cagsidy, P.S., Charlemagne, D., Preteseille, M. and Lelievre, L.G.:
Three Na ,K -ATPase forms in rat heart as revealed by K louabain
antagonism. FEBS Lett. 15_3_: 357-360, 1983.

Mansier, . d Lelievre, L.G.: Ca3+-free perfusion of rat heart reveals a
(Na +K )-ATPase form highly sensitive to ouabain. Nature ﬂit); 535-537,
1982.

Matsui, . agd Schwartz, A.: Mechanism of cardiac glycoside inhibition of the
(Na -K )—dependent ATPase from cardiac tissue. Biochim. Biophys. Acta
1_5;_: 655-663, 1967.

McCall, D.: Cation exchange and glycoside binding in cultured rat heart cells.

Mullins, L.J.: The generation of electric currents in cardiac fibers by Na/Ca
exchange. Am. J. Physiol. 233: C103-C110, 1979.

Nishizuka, Y.: The role of protein kinase C in cell surface signal transduction
and tumour promotion. Nature M: 693-698, 1984.

Noble, D.: Mechansm of action of therapeutic levels of cardiac glycosides.
Cardiovasc. Res. g: 495-514, 1980.

Noel, F. gnd podfraind, T.: Heterogeneity of ouabain specific binding sites and
(Na +K )-ATPase inhibition in microsomes from rat heart. Biochem.
Pharmacol. ﬂ: 47-53, 1984.

Onji, T. and Liu, M-S.: Ouabain receptor in isolated adult dog heart myocytes.
Arch. Biochem. Biophys. 201: 148-156, 1981.

Park, M.K. and Vincenzi, P.F.: Rate of onset of cardiotonic steroid-induced
inotropism: Inﬂuence of temperature and beat interval. J. Pharmacol. Exp.
Ther. 1_9§: 140-150, 1975.

Pollack, L.R., Tate, E.H. and ook, J.S.: Na+,K+-ATPase in HeLa cells after
prolonged growth in low K or ouabain. J. Cell. Physiol. m: 85-97, 1981.

Post, R.L., Hegyvary, C. and Kume, 8.: Activation by adenosine triphosphate in
the phosphorylation kinetics of sodium and potassium ion transport adeno-
sine triphosphatase. J. Biol. Chem. 251: 6530-6540, 1972.

Powis, D.A., And rs n, T.A., Jackson, H. and Wattus, G.D.: Stimulation of
neuronal Na ,K -ATPase by calcium. Biochem. Pharmacol. 3_2: 1219-1227,
1983.

Preuner, J.: Cardiac glycoside-induced changes of the Ca-binding in guinea pig
plasmalemma. Naunyn-Schmiedeberg's Arch. Pharmacol. m: R36, 1979.

Schwartz, A.: Is the cell membrane Na+,K+-ATPase enzyme system the
pharmacological receptor for digitalis? Circ. Res. g: 2-7, 1976.

Schwartz, A. and Adams, R.J.: Studies on the digitalis receptor. Circ. Res.
5_6_(Supp. 1): 1-154 - I-160, 1980.

132

Siegelbaum, S.A. and Tsien, R.W.: Calcium-activated transient outward current
in calf cardiac purkinje fibres. J. Physiol. 222: 485—406, 1980.

Silver, L.H. and Houser, S.R.: Transmembrane potassium ﬂuxes in isolated feline
ventricular myocytes. Am. J. Physiol. 242: H614-H621, 1985.

Solaro, R.J., Wise, R.M., Shiner, J.S. and Briggs, F.N.: Calcium requirements for
cardiac myofibrillar activation. Circ. Res. ﬂ: 525-530, 1974.

Temma, K. and Ak a, +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. 22_3: 490-496, 1982.

Tobin, T., Akera, T., Baskin, 8.1. and Brody, T.M.: Calcium ion and sodium- and
potassium-dependent adeonsine triphosphatase: Its mechanism of inhibition
and identification of the El-P intermediate. Mol. Pharmacol. _9_: 336-349,
1973.

Tobin, T., Akera, T., Han, 0.8. and Brody, T.M.: Lithium and rubidium
interactions with sodium- and potassium-dependent adenosine triphospha-
tase: A molecular basis for the pharmacological actions of these ions. Mol.
Pharmacol. 12: 501-508, 1974.

Wallick, E.T., Lanet UK. and Schwarts, A.: Regulation by vanadate of ouabain
binding to (Na ,K )-ATPase. J. Biol. Chem. 252: 8107-8109, 1979.

Ward, J.P.T. and Cameron, I.R.: Adaptation of the cardiac muscle sodium pump
to chronic potassium deficiency. Cardiovasc. Res. t2: 257-263, 1984.

Wellsmith, N.V. and Lindenmayer, G.E.: No receptor forms for ouabain in
sarcolemma-enriched preparations from canine ventricle. Circ. Res. ﬂ:
710-720, 1980.

Werdan, K., Wagenknecht, B. Zwissler, B., Brown, L., Krawietz, W. and Erdmann,
B.: Cardiac glycoside receptors in cultured heart cells - I. Characteriza-
tion of one single class of high affinity receptors in heart muscle cells from
chick embryos. Biochem. Pharmacol. _32: 55-70, 1984.

Werdan, K., Zwissler, B., Wagenknecht, B., Krawietz, W. and Erdmann, E.:
Quantitative correlation of cardiac glycoside binding to its receptor and

inhibition of the sodium pump in chicken heart cells in culture. Biochem.
Pharmacol. 22: 757-760, 1983.

Yamamoto, S., Akera, T. and Brody, T.M.: Sodium inﬂux rate and ouabain-
sensitive rubidium uptake in isolated guinea pig atria. Biochim. Biophys.
Acta ﬂ: 270-284, 1979.

Yamamoto, S., Akera, T. and Brody, T.M.: Positive inotropic action of
digoxigenin and sodium pump inhibition: Effects of enhanced sodium inﬂux.
J. Pharmacol. Exp. Ther. 2Q: 105-109, 1980.

Yingst, D.R.: Hemolysate increases calcium-inhibition of the Na+,K+ pump of
resealed human red cell ghosts. Biochim. Biophys. Acta m: 312-315, 1983.

133

Yingst, D.R. gnd Marcovitz, M.J.: Effect of hemolysate in calcium inhibition of
the (Na +K )-ATPase of human red blood cells. Biochem. Biophys. Res.
Commun. t1_1_: 970-979, 1983.

Yingst, D.R. and Polasek, P'Mt‘ Sensitivity and reversibility of Ca—dependent
inhibition of the (Na +K )-ATPase of human red blood cells. Biochim.
Biophys. Acta 8_1_§: 282-286, 1985.

Yoda, S. and Yoda, A.: ADP-and K+—sensitive phosphorylated intermediate of
Na,K-ATPase. J. Biol. Chem. 262: 1147-1152, 1986.

Zachowski, A., Lelievre, L., Aubry, J., Charlemagne, D. and Paraf, A.: Rgles+of
proteins from We” face of plasma membranes in susceptibility of (Na +K )—
stimulated Mg aenosine triphosphatase to ouabain. Proc. Natl. Acad. Sci.
USA ﬂ: 633-637, 1977.

Zavecz, J.H.: Intgragtion between theophylline and ouabain in the rabbit heart:
Effect on (Na K )-ATPase. Eur. J. Pharmacol. 22_0: 363-366, 1986.