A Possible Molecular Mechanism of the Action of
Digitalis
Ouabain Action on Calcium Binding to Sites Associated with a
Purified Sodium-Potassium-Activated Adenosine Triphosphatase
from Kidney
ANDRE GERVAIS, LOIS K. LANE, BEATRICE M. ANNER,
GEROGE E. LINDENMAYER, AND ARNOLD SCHWARTZ
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
SUMMARY Calcium binding at 0°C to a purified sheep kidney
Na + ,K + -ATPase was described by linear Scatchard plots. Binding
at saturating free calcium was 65-80 nmol/mg of protein, or 3040 mol of calcium/mol of enzyme. Aqueous emulsions of lipids
extracted from Na + ,K + -ATPase yielded dissociation constants
and maximum calcium-binding values that were similar to those
for native Na + ,K + -ATPase. Phospholipase A treatment markedly
reduced calcium binding. Pretreatment of native Na + ,K + -ATPase
with ouabain increased the dissociation constant for calcium binding from 131 ± 7 to 192 ± 7 /AM without altering maximum
calcium binding. Ouabain pretreatment did not affect calcium
binding to extracted phospholipids, ouabain-insensitive ATPases,
or heat-denatured Na + ,K + -ATPase. Na + and K + (5-20 mM) in-
creased the dissociation constants for calcium, which suggests
competition between the monovalent cations and calcium for the
binding sites. At higher concentrations of monovalent cations,
ouabain increased the apparent affinity of binding sites for
calcium. Extrapolation to physiological cation concentrations revealed that the ouabain-induced increase in apparent affinity for
calcium may be as much as 2- to 3-fold. These results suggest: (1)
calcium binds to phospholipids associated with Na 1 ,K-ATPase;
(2) ouabain interaction with Na + ,K + -ATPase induces a perturbation that is transmitted to adjacent phospholipids, altering
their affinity for calcium; and (3) at physiological concentrations
of Na * or K'. or both, ouabain interaction with Na + ,K'-ATPase
may lead to an increased pool of membrane-bound calcium.
THE SODIUM-POTASSIUM pump and its in vitro manifestation, sodium-potassium-activated adenosine triphosphatase (Na+,K+-ATPase), react with and are inhibited by
cardiac glycosides. It seems reasonable that such interaction is responsible for the positive inotropic effect of glycosides on the heart. Several hypotheses have been proposed
to explain how binding of cardiac glycosides to the
Na+,K+-ATPase causes intracellular responses. All are
based on the assumption that cardiac glycosides induce a
positive inotropic effect by raising intracellular calcium
concentration secondary to an interaction at the cell membrane.2 One popular hypothesis suggests that a membrane
system couples an exchange of sodium for calcium and that
inhibition of the sodium-potassium pump increases intracellular sodium;2""5 this, in turn, activates the membrane
exchange system and results in increased calcium influx or
decreased calcium efflux. Langer5 has extended this hypothesis to other inotropic interventions such as the Bowditch phenomenon and hypothermia. Although this hypothesis remains attractive, the universality of the concept
has been questioned; verapamil blocks the Bowditch response under conditions in which the response to ouabain
appears normal." Another hypothesis is that the glycosideNa+,K+-ATPase interaction may stabilize a conformation
of the pump that favors the exchange of internal potassium
for external calcium through the glycoside-pump complex.7 A third hypothesis is that glycosides affect a potassium-calcium exchange occurring during the cardiac action
potential.8 The latter two hypotheses have been challenged by a recent report showing a dissociation of the
positive inotropic response to cardiac glycosides from net
potassium loss.9
Calcium has direct effects on Na+,K+-ATPase. It inhibits activity through (1) formation of a complex with
ATP, (2) reduction in the apparent affinity of a site on the
enzyme for magnesium, and (3) reduction in the apparent
affinity for sodium with respect to sodium activation.2
Calcium may reduce the amount of potassium required for
half-maximal activation of Na+,K+-ATPase7 and, at very
low concentrations, may also "substitute" for sodium in
stimulating the binding of ouabain to Na+,K+-ATP-
From the Departments of Cell Biophysics and Medicine, Baylor College
of Medicine, and the Fondren-Brown Cardiovascular Research and Training Center of the Methodist Hospital, Houston, Texas 77030
Supported by HL 07906, HL 17269 (the Cell Membrane Section of the
National Heart and Blood Vessel Research and Demonstration Center,
Baylor College of Medicine, a grant-supported research project of the
National Heart, Lung, and Blood Institute, National Institutes of Health),
NIH Contract 71-2493, and grants from the American Heart Association,
Texas Affiliate, Houston Chapter.
Dr. Gervais is a Research Fellow, Conseil de la Recherche en Same du
Quebec, Canada, 760235; selected as finalist, American College of Cardiology, Young Investigator Award, 1976. Dr. Lane is a USPHS Postdoctoral Trainee, HL 02102. Dr. Anner is a Fogarty International Fellow,
USPHS, FO5 TWO2160-02. Dr. Lindenmayer is a recipient of an Established Investigatorship Award from the American Heart Association; his
present address is: Department of Pharmacology, Medical University of
South Carolina, Charleston, South Carolina 29401.
A preliminary report of the data was presented at the FASEB meeting
in Atlantic City in April, 1975.'
Address for reprints: Dr. Arnold Schwartz, Department of Cell Biophysics, Baylor College of Medicine, Houston, Texas 77030.
Received February 17, 1976; accepted for publication July 8, 1976.
OUABAIN ON Ca BINDING TO Na+,K+-ATPase/Ge/-va;s et al.
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
a s e 10.11 -f^ese effects, while complex, do suggest that
calcium can react with sites on the enzyme per se or with
sites closely associated with Na+,K+-ATPase in its native
lipoprotein (i.e., protein-membrane) configuration. On
the basis of such considerations, Besch and Schwartz12
suggested that the inotropic action of digitalis glycosides
may be due to a conformational change in Na + ,K + -ATPase such that sites associated with the system would have
increased affinity for calcium. In effect it was suggested
that glycoside interaction with Na + ,K + -ATPase induces an
increase in a pool of calcium bound to the cell membrane.
Also, inherent in their proposal was the concept that the
number of calcium-binding sites could be greater than, for
example, the number of sodium-activation sites (i.e.,
thought to be three per molecule of enzyme2). This hypothesis could not be tested at the time of its proposal,
because of the impurity of Na + ,K + -ATPase preparations.
With the availability of purified Na + ,K + -ATPases, 13 the
concept can be examined, and the results presented herein
confirm the suggestion12 that cardiac glycoside interaction
with Na+,K+-ATPase modulates the affinity for calcium of
sites on the purified enzyme preparation.
Methods
ENZYME ISOLATION
Highly active Na + ,K + -ATPase preparations were isolated from the outer medulla of frozen lamb kidneys by the
method of Lane et al.14 Na + ,K + -ATPase activities were
measured at 37°C by a spectrophotometric, coupled-enzyme assay in a medium containing histidine, 25 ITIM, pH
7.2; Na2ATP, 2.5 miu; MgCl2, 5 miu; NaCl, 100 miu;
KG, 10 miu; tris(hydroxymethyl)aminomethane-ethylenediaminetetraacetic acid (Tris-EDTA), 1 min; NADH,
0.40 HIM; phosphoenolpyruvate, 2 miu; and pyruvate kinase/lactic acid dehydrogenase (Sigma). Protein concentrations of these preparations were determined by the
method of Lowry et al.,15 using bovine serum albumin as
the standard. Specific activities ranged from 900 to 1,200
/imol of ATP hydrolyzed per mg of protein per hour.
Maximal ouabain binding, using [3H]ouabain, was carried
out in the presence of ATP, magnesium, and sodium, or
magnesium and inorganic phosphate, as previously described.'"
Ca2+ BINDING TO Na+,K+-ATPase PREPARATION
Assays were carried out at 0°C in the presence of imidazole, 25 mM, pH 7.3; CaCl2, 10-1,000 JAM, containing
[•^Ca], 0.4 mCi/mmol was added, in a final volume of 2
ml. The radioactivity in each tube varied proportionally to
the concentration of CaCl2. The reaction was initiated by
the addition of 200 /j.g (protein) of Na + ,K + -ATPase, and
the tubes were immediately centrifuged at 105,000 g for
15 minutes.
The supernatant fraction was removed, the pellet was
solubilized with NaOH, added to a scintillation medium,
and assayed for radioactivity. This yielded total radioactivity in the pellet. Radioactivity in the water space of the
pellets was determined by separate assays which included
the amount of [45Ca] normally present in assays with 10
/J.M CaCl2 for total binding plus additional unlabeled
CaCl2 at 1 or 10 miu. The counts per minute (cpm) found
in this pellet were equivalent to radioactivity not bound to
the specific binding sites but trapped by the pellet or
associated with the water space. The radioactivity contained in the water space for each concentration of CaCl2
was calculated, assuming that the amount of radioactivity
in the water space was proportional to the cpm added. The
cpm calculated for each concentration was subtracted from
the total amount to obtain the net radioactivity, which was
taken to represent calcium bound to the membrane preparation. Equivalent values for the isotope content in the
water space were obtained with 1 and 10 miu unlabeled
calcium, and these values were similar to those obtained in
independent experiments by the use of [22Na]. We assumed that the amount of isotope contained in any fraction of solvent is independent of calcium concentration,
whereas the amount of isotope bound to the enzyme is a
function of the specific activity. Endogenous calcium in
the enzyme preparation was determined by atomic absorption spectroscopy of HCl-LaCl,-, extracts of the enzyme
preparation and buffer solutions. The amount was usually
1-2 /iM per assay. The values for total and water space
were unaffected by longer incubation or centrifugation
times.
PREPARATION OF A OUABAIN-Na+,K+-ATPase
COMPLEX
The Na + ,K + -ATPase preparation (about 3 mg) was incubated with 1.8 miu ouabain in the presence of imidazole, 25 miu, pH 7.4, and MgCl2, 1.1 miu, for 30 minutes at
room temperature. This condition yields maximum ouabain binding to the Na+,K+-ATPase.1(i Binding was promoted by magnesium alone in these experiments, because
other binding conditions contained ligands that complex
with calcium, thereby reducing free calcium concentrations in the assays for calcium binding. Controls were
treated in an identical way except that ouabain was absent.
Subsequently, samples were taken to measure calcium
binding as described above. The final magnesium concentration in the calcium binding assays was 57 /AM.
EXTRACTION OF L1PIDS FROM Na+,K+-ATPase
PREPARATIONS
Lipids were extracted with chloroform-methanol (2:1).
After removal of the solvent by rotary evaporation, the
lipids were redispersed in Tris-EDTA, 1 miu, by homogenization, and total lipid phosphate was determined by the
method of Bartlett.17 Samples were taken to measure
calcium binding as described above.
PHOSPHOLIPASE A TREATMENT OF THE Na\K + ATPase
Phospholipase A was partially purified from Crotalus
adarnanteus venom and reacted with the Na + ,K + -ATPase
preparations. The reaction consisted of a 10-minute incubation at 37°C in the presence of the Na + ,K + -ATPase (25
mg protein), phospholipase A (5 mg), CaCl2 (5 miu),
imidazole (0.25 M, pH 7.2), and 1% fatty acid-poor bovine serum albumin. The reaction was terminated by dilution and centrifugation followed by multiple washes. Controls were treated in a similar manner except for the
absence of the phospholipase A.
CIRCULATION RESEARCH
10
cpm x I0"
20
1.87
Added
18.7
—|—
187
•10
VOL. 40, No.
1, JANUARY
1977
concentration extremes, since in the lower range contaminating (endogenous) calcium became significant, while in
the upper range, [45Ca] in the water space comprised most
of the [45Ca] in the pellet (Fig. 1). The amount of binding
was unaffected by longer incubation times or longer centrifugation times.
THE EFFECT OF OUABAIN ON CALCIUM BINDING TO
THE ATPase PREPARATION
100
1000
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
[CaCI 2 l p M l ]
FIGURE 1 Calcium binding to a purified Na+, K+-A TPase preparation from the outer medulla of sheep kidney. The assays were
carried out at 0°C in the presence of imidazole, 25 mwi, pH 7.3,
CaCI2, and /45CaJ as shown. The reactions were started by addition
of 200 fj.g (protein) of the membrane preparation (total volume =
2 ml). Subsequently, the membranes were collected in pellets and
assayed for [4iCaJ content. "Water Space" and "Net" cpm were
calculated as described under Methods. Results depicted in Figures
1-7 are individual experiments which are representative of four to
eight experiments.
HEAT-DENATURED Na + ,K+-ATPase PREPARATIONS
AND ISOLATION OF GLYCOSIDE-INSENSITIVE ATPase
PREPARATIONS
We took advantage of the fact that ouabain interacts
with a site on the Na+,K+-ATPase in a pseudoirreversible
manner2 to prepare a ouabain-membrane complex that
subsequently could be employed in calcium-binding assays. Magnesium was used to stimulate ouabain binding in
these experiments, and a high ouabain concentration (1.8
miu) was employed to ensure that the receptor sites on the
Na+,K+-ATPase were saturated with the drug. The magnesium concentration in the assays for calcium binding was
57 /XM, which increased the dissociation constant for calcium binding to the control preparation (i.e., not exposed
to ouabain) by 1.3-fold. Ouabain pretreatment increased
the dissociation constant for calcium binding from 131 /XM
to 192 /XM without changing the number of calcium-binding sites (i.e., 68 and 71 nmol/mg of protein for control
and drug-treated preparations, respectively) (Fig. 2 and
Table 1 A). Experiments carried out at 22°C with sheep or
dog kidney as sources for the ATPase yielded similar
results (Tables 2 and 3). Prednisolone was used to test the
possibility that the ouabain effect was due to the steroid
nature of the compound rather than to a reaction with a
specific cardiac glycoside receptor site on or in the
The Na+,K+-ATPase was denatured by placing it in a
boiling water bath for 5 minutes; this resulted in a loss of
catalytic activity without coagulation of the preparation.
Subsequently, the preparation was resuspended by hand
with a glass homogenizer fitted with a Teflon pestle. Glycoside-insensitive ATPases employed were those present
in isolated cardiac sarcoplasmic reticulum18 and in mitochondria.19
Results
CALCIUM BINDING TO SHEEP KIDNEY Na + ,K + -ATPase
PREPARATIONS
Calcium, in a concentration range of 10-1,000 /XM,
adsorbed to the sheep kidney preparations in a manner
suggesting saturation of the adsorption sites at higher
concentrations of free calcium (Fig. 1). A Scatchard plot
of the data revealed that adsorption occurs, at least predominantly, to a single species of noninteracting sites on
the membrane preparation (Fig. 2), with a dissociation
constant (KD) for the adsorption process of approximately
100 /XM (i.e., in the absence of magnesium; see below).
Extrapolation to the abscissa suggested that the number of
calcium-binding sites in the preparations was 50-70 nmol/
mg of protein. In other experiments (not shown) which
employed free calcium concentrations of less than 10 /J.M
or greater than 1,000 /XM, there was evidence of additional calcium-binding species in the preparation. The
reliability of the binding technique employed in the present study, however, was significantly reduced at these
FIGURE 2 Scatchard plots of calcium binding to control (O),
oeabain-pretreated ( • ) and prednisolone-pretreated (A) Na+,K+A TPase preparations. Pretreatment of approximately 3 mg protein
of the Na+,K+- ATPase preparation was carried out at room temperature for 30 minutes in the presence of imidazole, 25 miu, pH
7.4; MgCl2, 1.1 mM, and, where indicated, ouabain, 1.8 rain, or
prednisolone sodium succinate, 1.8 mu. Subsequently, the control
and drug-treated preparations were used for calcium-binding experiments as described in Figure 1 and Methods. Final magnesium
and drug (where present) concentrations in the assays for calcium
binding were 57 and 93 /XM, respectively. Least squares analysis
was used to determine the lines presented.
OUABAIN ON Ca BINDING TO Na+,K+-ATPase/Gervaw et al.
TABLE 1 Effects of Ouabain Pretreatment on Calcium
Binding to Purified Sheep Kidney Na+,K+-ATPase
Preparations at 0°C
KD(> M )
Bma,
(nmol/mg)
Experiment no.
Control
Ouabain
Control
Ouabain
4
5
6
7
8
146
139
116
131
130
121
165
99
188
197
162
219
196
212
201
164
70
71
57
64
59
60
91
69
70
73
65
75
67
70
88
63
Mean
131
192
7
68
4
71
3
1
2
3
SEM
7
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
Results from eight experiments carried out as in Figure 2. Least squares
analysis of each data set was used to calculate Kn and BmaK.
TABLE 2 Effects of Ouabain Pretreatment on Calcium
Binding to Purified Sheep or Dog Kidney Na+,K+-A TPase
Preparations in the Absence of Sodium at 22°C
K D ( >«)
Experiment no.
B m .x
(nmol/mg)
Control
Ouabain
Control
Ouabain
1
2
3
100
93
115
155
98
157
65
59
75
61
53
66
Mean
103
7
137
19
66
5
60
4
96
83
70
111
128
72
91
84
73
67
87
61
83
8
104
17
83
5
72
Sheep Kidney
Na+,K+-ATPase
SEM
Dog Kidney
Na+,K+-ATPase
1
2
3
Mean
SEM
8
The enzyme was pretreated with 1 0 * M ouabain and 1.5 MM Mg in 25
mM imidazole, centrifuged for 20 minutes at 100,000g and, after removal
of the supernatant fraction, the pellets were resuspended in 25 mM imidazole for calcium binding. The temperature was maintained at 22°C during
the entire experiment.
Na+,K+-ATPase. Pretreatment with prednisolone, however, had no effect on calcium binding (Fig. 2). Ouabain
was still effective at a much lower concentration (10 /XM),
and the effect persisted after essentially all of the unbound
drug was removed from the medium containing the membrane fragments (Table 4). Furthermore, ouabain had no
effect on calcium binding to a heat-denatured Na+,K+ATPase preparation (Fig. 3), on calcium binding to lipids
extracted from the Na+,K+-ATPase and subsequently redispersed in an aqueous medium (Fig. 4), or on calcium
binding to mitochondrial and sarcoplasmic reticulum preparations which contain glycoside-insensitive ATPase activities (data not shown). These results supported the conclusion that the ouabain effect on calcium binding is a consequence of the drug reacting with the Na+,K+-ATPase
protein.
THE EFFECT OF MONOVALENT, DIVALENT, AND
TRTVALENT CATIONS ON CALCIUM BINDING TO THE
Na+,K-ATPase
Sodium (Fig. 5 and Table 3) and potassium (similar to
sodium, data not shown) appeared to compete with calcium for the binding sites, at least in a concentration range
of 5-20 mM. Pretreatment with ouabain markedly affected
the ability of sodium to antagonize calcium binding. Ouabain increased the "inhibitory constant" for sodium from 9
to 41 mM and had a similar effect on potassium as an
inhibitor of calcium binding. Thus, ouabain reduced the
affinity for sodium by 78% (Fig. 5), whereas it reduced
the affinity for calcium in the absence of sodium by only
32%. This difference led to the conclusion that ouabain
increased the affinity of the sites for calcium when high
sodium (or potassium) was present. At 20 mM sodium, a
1.45-fold increase in affinity for calcium was observed
(Fig. 5). Extrapolation to a physiological concentration of
sodium (e.g., 140 mM) suggested that ouabain may induce
as much as a 2- to 3-fold increase in the affinity for
calcium. Sodium and potassium also appeared to decrease,
to some extent, the number of calcium-binding sites (inset
of Fig. 5).
In other studies, 0.1-10 mM magnesium and 10 /XM
lanthanum were found to have a mixed effect on calcium
binding by increasing the KD and decreasing the number of
binding sites; but 10 /IM verapamil, a potent calcium
antagonist with respect to myocardial contractility,20 had
no effect (Fig. 6).
TABLE 3 Effect of Ouabain Pretreatment on Calcium
Binding to Purified Sheep Kidney Na+,K+-ATPase
Preparations in the Presence of 20 mM Sodium at 22"C
(nmol/mg)
KD ( M M )
Experiment no.
Control
Ouabain
Control
Ouabain
1
2
3
382
264
414
173
121
116
61
52
78
38
32
28
Mean
353
46
137
18
64
33
3
SEM
The procedure was the same as in Table 2, except that 20 mM sodium
chloride was added to the tubes in the calcium-binding experiments.
TABLE 4 Effect of Lower Ouabain Concentrations on
Calcium Binding to Purified Sheep Kidney Na+,K+A TPase Preparations
Condition
Control
10 ptM ouabain
Control
10 fj.M ouabain plus
wash
KD ( M M )
122
197
102
169
±
±
±
±
14
11
20
23
Bmax (nmol/mg)
73
88
64
72
±
±
±
±
4
3
7
7
These experiments (n = 6) we're carried out as described in Figure 2 with
two exceptions. In both, 10 MM instead of 1.8 mM ouabain was used in the
pretreatment stage. In the second set, bound but not free ouabain remained after centrifugation and resuspension of the ouabain-membrane
complex. (Parallel experiments using [3H|ouabain were used to document
this statement.) KD and B ^ x were determined from least squares analysis
of the data sets. Mean values ± SD are presented.
CIRCULATION RESEARCH
12
VOL. 40, No. 1, JANUARY 1977
amount of phosphatidylserine, which carries a net negative
charge and, as a consequence, could serve as a binding site
for calcium. In an attempt to identify the adsorption site
for calcium, interaction of calcium with extracted phospholipids was measured. Binding to these lipids was described by adsorption to a single species of binding sites
characterized by a dissociation constant of 147 YM and
maximal binding of 86 nmol (Fig. 4). This compares favor-
a 03,-
, 0.02
; a oi
80,
i
0—^_.
2
4
6
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
FIGURE 3 Scatchard plots of calcium binding to heat-denatured
preparations with ( • ) and without (O) ouabain
Na^,K+-ATPase
pretreatment. This experiment was carried out as described in
Figure 2 except that the preparation was placed in a boiling water
bath for 5 minutes prior to the pretreatment stage ( ± ouabain). The
exposure to heat destroyed catalytic activity of the Na+,K*-A TPase
but did not result in visible clumping of the denatured membrane
fragments. The preparation was gently homogenized by hand with
a small glass homogenizing vessel and Teflon pestle after heating.
1
10
NaCI I mM)
-20
0.06
•°
•\p
0
2
4
6
o
5
I 40 -
Bound C a 2 4 I p M )
0.02
Cr
8
"max
0
8
Bound Ca?* (uM I
FIGURE 4 Scatchard plots of calcium binding to lipids extracted
from the Na+,K+-ATPase preparations with ( • ) and without (O)
ouabain pretreatment of the lipids. Extraction of lipids and redispersal in an aqueous medium were carried oul as described under
Methods. Pretreatment ± ouabain and subsequent assays for calcium binding were carried out as described in Figure 2.
i
1/
FIGURE 5 Effects of sodium on the apparent dissociation constant (K'D) and maximal number of sites (Bmax) for calcium binding to control and ouabain-pretreated Na+,K+-ATPase preparations. A series of binding assays was carried out and analyzed as
described in Figure 2 except that NaCI, 0, 5, 10, or 20 mM, was
present in the assays. Least squares analysis was used to determine
KD and Bmaxfor the various sodium concentrations. Plots ofK,, vs.
sodium and Bmax vs. sodium (inset) for preparations with (%)
and without (O) ouabain pretreatment are presented. Each point is
the average of two determinations. Extrapolation to the abscissa
yielded negative values of dissociation constants for sodium. Similar responses were obtained when potassium was substituted for
sodium, and ouabain pretreatment increased the "inhibitory constant" for potassium from 9 to 27 mM.
100 r
80
60
NATURE OF THE ADSORPTION SITES FOR CALCIUM
The number of calcium binding sites is 25-40 times
greater than the number of Na+,K+-ATPase molecules in
the preparation, as measured by ouabain-binding experiments. There are at least three possibilities as to the
chemical nature of these sites. The sites could be (1) the
negatively charged amino acid residues on the external
surfaces of the polypeptides, (2) the glycoprotein which
contains negatively charged sialic acid residues,13 or (3)
membrane lipids. Recent experiments on the same
Na+,K+-ATPase preparations that were employed for
these studies showed that the preparations contained 133266 mol of lipid phosphate per mol of ouabain-binding
sites. Other studies revealed the presence of a significant
40
20
0.1 mM 1.0 mM
'
n
10 mM
w
IOUM
hich
lO^M
Verapamil
FIGURE 6 Effects of magnesium, lanthanum, and verapamil on
calcium binding to the Na+,K*-ATPase preparation. Calcium
binding was carried oul in the presence of 250 /IM free calcium as
described in Figure 1. Concentrations of the modifiers in the assays
were as shown.
OUABAIN ON Ca BINDING TO Na+,K+-ATPase/Gerva« et al.
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
ably with the values obtained for the native preparation
(Table I). (Equivalent amounts of phospholipids were
present in assays of the two preparations.) These similarities strongly suggested that calcium was binding to the
membrane lipids in the Na+,K+-ATPase preparation.
The result of a second experiment also was consistent
with the conclusion that calcium was adsorbing to the
membrane phospholipids. Phospholipase A treatment of
the Na+,K+-ATPase preparation was found to reduce the
phospholipid content and Na+,K+-ATPase activity to 40%
and 28%, respectively, of control values. The treated
preparation, however, retained the ability to bind ouabain
(maximal binding was 99% of control) and to be phosphorylated by ATP in the presence of magnesium and sodium
(71% of control). The latter data are compatible with the
suggestion that significant native configurations of the
enzyme are retained after phospholipase A treatment.
Calcium binding to the treated preparations, however, was
markedly reduced compared to the native preparation
(Fig. 7).
Discussion
It generally has been recognized that in order to catalyze
the hydrolysis of ATP, the Na+,K+-ATPase must be associated with lipids or at least a lipid-like environment (e.g.,
some detergents).2 Arrhenius plots of Na+,K+-ATPase
activity are characterized by an inflection point between
15°C and 20°C; the plots become linear on partial removal
of the lipids and are reconverted to a nonlinear form when
phospholipids are added back to the preparations.21 The
discontinuity in the Arrhenius plot was found by calorimetric22 and electron spin resonance23 studies to be related to
temperature-induced transitions in the physical state of the
lipids. In molecular terms this means that the macromolecular structure of the Na+,K+-ATPase can be modified by
the composition and physical state of the surrounding
lipids.
= o.oi
I
0
1
2
3
4
5
Bound Ca 2 * < U M >
FIGURE 7 Scatchard plots of calcium binding to a native (O) and
a phospholipase A-treated (•) Na+,K+-ATPase preparation. Details of the treatment are presented under Methods. The native
preparation was pretreated in a manner identical to that for the
lipase-treat'ed preparation except for the absence of phospholipase
A.
13
The transfer of structural information from membrane
lipids to the Na+,K+-ATPase may also occur in the opposite direction (i.e., from enzyme to lipids). Grisham and
Barnett24 found that in a partially purified preparation the
phospholipid bilayer adjacent to the enzyme was more
ordered than that of the overall membrane but the interior
of the bilayer next to the Na+,K+-ATPase was more fluid.
Therefore, conformation changes in the Na+,K+-ATPase
could perturb the structure of its adjacent lipids. The
studies reported herein support these suggestions. Ouabain reacts with a specific site on the Na+,K+-ATPase,2
and Ruoho and Kyte25 have tentatively localized the receptor region on the higher molecular weight polypeptide
in the purified Na+,K+-ATPase preparations. In the present study, the ouabain effect on calcium binding was
shown to require the native integrity of the preparation.
Ouabain did not affect calcium binding to extracted phospholipids, and the effect on calcium binding and inhibition
of catalytic activity persisted after removal of unbound
drug. These characteristics are consistent with the conclusion that ouabain affects calcium binding by reacting with
a site on a protein moiety of the Na+,K+-ATPase. Furthermore, it had previously been concluded that glycoside
interaction with this receptor stabilized the Na+,K+-ATPase in a particular conformation.2 It is reasonable, therefore, to suppose that such an action would result in an
altered reactivity of sites on or associated with the
Na+,K+-ATPase.
The conclusion that the ouabain effect is transmitted
from protein to lipids, the latter serving as sites for calcium
binding, is less secure. The possibility that calcium binds to
negatively charged residues on certain surfaces (i.e., those
exposed to the aqueous environment) of the protein is not
eliminated by these studies. It is important to note that any
procedure required to remove all of the lipids in Na+,K+ATPase denatures the preparation.
Based on Kyte's analysis,26 it appears that the number of
sialic acid residues on the smaller polypeptide in the preparation is too small to account for the number of calciumbinding sites. On the other hand, several experiments
strongly suggest that calcium binding occurred to phospholipids in the preparation. First, there were more than
enough phospholipids in the preparation to account for
calcium binding and, at least in semiquantitative terms,
about 17% of the phospholipid was found to consist of
phosphatidylserine. This species has net negative charges
at pH 7.4 and, therefore, could serve as a site for calcium
binding. Second, lipid extracts bound calcium in a manner
similar to that of the native preparation. Third, phospholipase A treatment reduced the phospholipid content and
calcium binding while leaving the Na+,K+-ATPase with
considerable ability to react with ATP and ouabain.
Sodium and potassium inhibited calcium binding in a
similar manner, with "inhibitory" dissociation constants of
approximately 9 mM. Pretreatment with ouabain increased the constants to 41 mM for sodium and 27 mM for
potassium. These constants were much greater than those
of the sodium and potassium activation sites for sodium
and potassium, respectively,27'28 but were in the range of
those which characterize sodium and potassium interaction with phospholipids.29 Thus, the basis of the antago-
CIRCULATION RESEARCH
14
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
nism between monovalent cations and calcium may be a
competition for negatively charged headpieces of phospholipids in the bilayer around the Na+,K+-ATPase. An
alternate possibility is that the monovalent cations are
modifying the surface charge of the bilayer without actually binding to the negative charges in the sense visualized for calcium adsorption.30
Regardless of the nature of the calcium-binding site, the
present studies may pertain to the question of how the
digitalis glycosides produce an inotropic effect in the
heart. Glycoside interaction with the sodium-potassium
pump potentially could result in an increase in a membrane-bound pool of calcium. Attempts to extrapolate
these considerations to the action of cardiac glycosides on
the heart must be tempered by the fact that the studies
employed a highly purified Na+,K+-ATPase preparation
from kidney. Furthermore, if the results do apply to heart,
some as yet unknown mechanism is still required to move
the calcium from the membrane into the cell in a nonelectrogenic manner, since voltage clamp experiments have
yielded data confirming the lack of direct effect of digitalis
on a calcium current.31"33 It is therefore possible that an
electroneutral exchange diffusion system3"5 is operative in
moving calcium into the cell.
Acknowledgments
We thank Drs. Mark L. Entman, Earl T. Wallick, and Julius C. Allen
for helpful discussions, and Edward P. Bornet for isolating the sarcoplasmic reticulum preparations.
References
l. Gervais A, Lane LK, Lindenmayer GE: Ouabain (OU) effect on
calcium (Ca++) binding to purified sheep kidney Na+,K+-ATPase (E).
Fed Proc34: 715, 1975
2. Schwartz A, Lindenmayer GE, Allen JC: The sodium-potassium
adenosine triphosphatase; pharmacological, physiological and biochemical aspects. Pharmacol Rev 27: 3-134, 1975
3. Reuter H, Seitz NJ: The dependence of calcium efflux from cardiac
muscle on temperature and external ion composition. J Physiol (Lond)
195: 451-470, 1968
4. Baker PF, Blaustein MP, Hodgkin AL, Steinhardt RA: The influence
of calcium on sodium efflux in squid axons. J Physiol (Lond) 200:431458, 1969
5. Langer GA: The intrinsic control of myocardial contraction ionic
factors. N Engl J Med 285: 1067-1071, 1971
6. McCans JL, Lindenmayer GE, Munson RG, Evans RW, Schwartz A:
A dissociation of positive staircase (Bowditch) from ouabain-induced
positive inotropism. Circ Res 35: 439-447, 1974
7. Lindenmayer GE, Schwartz A: A kinetic characterization of calcium
on Na+,K+-ATPase and its potential role as a link between extracellular and intracellular events; hypothesis for digitalis-induced inotropism. J Mol Cell Cardiol 7: 591-612, 1975
8. Morad M, Greenspan AM: Excitation contraction coupling as a possible, site for the action of digitalis on heart muscle. In Cardiac Arrhythmias, edited by LS Dreifus, W Likoff. New York, Grune & Stratton,
1973, pp 479-489
9. Poole-Wilson PA, Langer GA: Glycoside inotropy in the absence of
an increase in potassium efflux in the rabbit heart. Circ Res 37: 390395, 1975
10. Schwartz A, Kirton CB, Kraft GH, Entman ML, Allen JC: Effect of
VOL. 40, No.
1, JANUARY
1977
low concentrations of calcium on H3-ouabain binding to Na+,K+ATPase. Fed Proc 31: 904, 1972
11. Schon R, Schonfeld W, Menke KH, Repke RH: Mechanism and role
of Na+/Ca++competition in Na+,K+-ATPase reactions. Acta Biol Med
Ger29: 643-659, 1972
12. Besch HR Jr, Schwartz A: On a mechanism of action of digitalis. J Mol
Cell Cardiol 1: 195-199, 1970
13. Jorgensen PL: Isolation and characterization of the components of the
sodium pump. Q Rev Bophys 7: 239-274, 1975
14. Lane LK, Copenhaver JH Jr, Lindenmayer GE, Schwartz A: Purification and characterization of and (3H]ouabain binding to the transport
adenosine triphosphatase from outer medulla of canine kidney. J Biol
Chem 248:7197-7200, 1973
15. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265-275, 1951
16. lnagaki C, Lindenmayer GE, Schwartz A: Effects of sodium and
potassium on binding of ouabain to the transport adenosine triphosphatase. J Biol Chem 249: 5135-5140, 1974
17. Bartlett GR: Phosphorous assay in column chromatography. J Biol
Chem 234: 466-468, 1959
18. Harigaya S, Schwartz A: Rate of calcium binding and uptake in
normal animal and failing human cardiac muscle. Circ Res 25: 781794, 1969
19. Lindenmayer GE, Sordahl LA, Schwartz A: Reevaluation of oxidative
phosphorylation in cardiac mitochondria from normal animals and
animals in heart failure. Circ Res 23: 439-450, 1968
20. Fleckenstein A: Specific inhibitors and promoters of calcium action in
the excitation-contraction coupling of heart muscle and their role in
the prevention of production of myocardial lesions. In Calcium and the
Heart, edited by P Harris, LH Opie. New York, Academic Press,
1971, pp 135-188
21. Charnock JS, Cook DA, Almeida AF, Rebecca T: Activation energy
and phospholipid requirements of membrane-bound adenosine triphosphatases. Arch Biochem Biophys 159: 393-399, 1973
22. Kimelberg HK, Papahadjopoulos D: Effects of phospholipid acyl
chain fluidity, phase transition and cholesterol on (Na+ + ^^stimulated adenosine triphosphatase. J Biol Chem 249: 1071-1080, 1974
23. Grisham CM, Barnett RE: The role of lipid-phase transitions in the
regulation of the (sodium + potassium) adenosine triphosphatase.
Biochemistry 72: 2634-2635, 1973
24. Grisham CM, Barnett RE: The interrelationship of membrane and
protein structure in the functioning of the (Na+ + Reactivated
ATPase. Biochim Biophys Acta 266: 613-624, 1972
25. Ruoho A, Kyte J: Photo-affinity labeling of the ouabain-binding site
on (Na+ + K+)-adenosine triphosphatase. Proc Natl Acad Sci USA
71:2352-2356, 1974
26. Kyte J: Properties of the two polypeptides of sodium and potassiumdependent adenosine triphosphatase. J Biol Chem 247: 7642-7649,
1972
27. Lindenmayer GE, Schwartz A, Thompson HK Jr: A kinetic description for sodium and potassium effects on (Na+ + K+)-adenosine
triphosphatase; a model for a two-nonequivalent site potassium activation and an analysis of multiequivalent site models for sodium activation. J Physiol (Lond) 236: 1-28, 1974
28. Garay RP, Garrahan PJ: The interaction of sodium and potassium
with the sodium pump in red cell. J Physiol (Lond) 231: 297-325,
1973
29. Barton PG: The influence of surface charge density of phosphatides on
the binding of some cations. J Biol Chem 243: 3884-3890, 1968
30. Jones MN: The electrical double layer. In Biological Interfaces. New
York, Elsevier, 1975, pp 135-161
31. McDonald TF, Nawrath H, Trautwein W: Membrane currents and
tension in cat ventricular muscle treated with cardiac glycosides. Circ
Res 37: 674-682, 1975
32. Greenspan AM, Morad M: Electromechanical studies on the inotropic
effects of acetylstrophanthidin on ventricular muscle. J Physiol (Lond)
253: 357-384, 1975
33. Beeler GW Jr: Ionic currents in cardiac muscle; a framework for
glycoside action (FASEB Symposium, New Aspects of Cardiac Glycoside Action, April, 1976). Fed Proc 1977 (in press)
A possible molecular mechanism of the action of digitalis: ouabain action on calcium binding
to sites associated with a purified sodium-potassium-activated adenosine triphosphatase
from kidney.
A Gervais, L K Lane, B M Anner, G E Lindenmayer and A Schwartz
Downloaded from http://circres.ahajournals.org/ by guest on July 28, 2017
Circ Res. 1977;40:8-14
doi: 10.1161/01.RES.40.1.8
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1977 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/40/1/8
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/