Ouabain Binding and Cation Transport in Human Erythrocytes

Abstract

In the present studies we have explored the relation between ouabain binding and the inhibition of potassium influx in intact human erythrocytes. The rate at which bound ouabain molecules dissociate from the erythrocyte membrane is not altered by complete replacement of choline with sodium or by partial replacement with potassium. These findings indicate that the effects of these cations on ouabain binding reflect alterations in the rate of association of ouabain molecules with the erythrocyte membrane. Variations in the cation composition of the incubation solution did not alter the relation between the fraction of the glycosidebinding sites occupied by ouabain or the fraction of ouabain-sensitive potassium influx which was inhibited. That is, irrespective of the affinity of the erythrocyte membrane for ouabain molecules and irrespective of the magnitude of glycoside-sensitive potassium influx, occupation of a given fraction of the glycoside-binding sites by ouabain results in the inhibition of an equal fraction of the ouabain-sensitive potassium transport sites.

Full text

Ouabain
Binding
and
Cation
Transport
in
Human
Erythrocytes
JRmRY
D.
GARDN
and
DiANE
R.
KINo
From
the
Digestive
and
Hereditary
Diseases
Branch,
National
Institute
of
Arthritis,
Metabolism
and
Digestive
Diseases,
National
Institutes
of
Health,
Bethesda,
Maryland
20014
A
B
S
T R
A
C
T
In
the
present
studies
we
have
explored
the
relation
between
ouabain
binding
and
the
inhibition
of
potassium
influx
in
intact
human
erythrocytes.
The
rate
at
which
bound
ouabain
molecules
dissociate
from
the
erythrocyte
membrane
is
not
altered
by
complete
replacement
of
choline
with
sodium
or
by
partial
re-
placement
with
potassium.
These
findings
indicate
that
the
effects
of
these
cations
on
ouabain
binding
reflect
alterations
in
the
rate
of
association
of
ouabain
mole-
cules
with
the
erythrocyte
membrane.
Variations
in
the
cation
composition
of
the
incubation
solution
did
not
alter
the
relation
between
the
fraction
of
the
glycoside-
binding
sites
occupied
by
ouabain
or
the
fraction
of
ouabain-sensitive
potassium
influx
which
was
inhibited.
That
is,
irrespective
of
the
affinity
of
the
erythrocyte
membrane
for
ouabain
molecules
and
irrespective
of
the
magnitude
of
glycoside-sensitive
potassium
influx,
oc-
cupation
of
a
given
fraction
of
the
glycoside-binding
sites
by
ouabain
results
in
the
inhibition
of
an
equal
frac-
tion
of
the
ouabain-sensitive
potassium
transport
sites.
INTRODUCTION
Using
intact
human
erythrocytes,
we
have
previously
explored
the
first
step
in
the
mechanism
of
action
of
cardioactive
glycosides;
namely,
the
binding
of
ouabain
to
the
plasma
membrane
(1).
These
studies
indicated
that
ouabain
binding
to
the
erythrocyte
membrane
is
specific,
reversible,
involves
a
single
class
of
binding
sites,
can
be
detected
at
ouabain
concentrations
as
low
as
10'
M,
and
correlates
directly
with
inhibition
of
erythrocyte
potas-
sium
influx.
Furthermore,
ouabain
binding
appears
to
involve
a
combination
of
glycoside
molecules
with
a
membrane
"receptor"
composed
of
a
glycoside-binding
site
and
a
cation
site,
and
the
number
and
type
of
ca-
Received
for
publication
8
November
1972
and
in
revised
form
16
March
1973.
tions
occupying
the
cation
site
determine
the
affinity
of
the
glycoside-binding
site
for
glycoside
molecules.
Much
of
our
understanding
of
membrane
cation
trans-
port
has
come
from
studies
of
the
effects
on
ion
transport
of
altering
the
cation
composition
of
the
incubation
me-
dium
(2).
These
studies
frequently
involve
comparing
the
effects
of
a
particular
alteration
to
those
of
the
same
alteration
in
the
presence
of
ouabain
(3,
4).
Since
al-
tering
the
cation
composition
of
the
incubation
medium
also
alters
the
apparent
affinity
with
which
ouabain
is
bound
to
the
cell
membrane
(1),
our
understanding
of
the
mechanism
by
which
cardioactive
glycosides
alter
ion
transport
would
be
facilitated
by
distinguishing
effects
of
cations
on
ion
transport
from
their
effects
on
ouabain
binding.
In
the
present
study
we
have
explored
the
events
which
occur
subsequent
to
ouabain
being
bound
to
the
cell
membrane;
namely,
the
inhibition
of
potas-
sium
influx.
In
particular,
we
have
examined
the
relation
between
the
amount
of
ouabain
bound
to
the
erythro-
cyte
membrane
and
the
magnitude
of
the
inhibition
of
potassium
influx.
We
have
also
explored
the
effects
of
cations
on
each
of
these
two
phenomena
as
well
as
on
the
relation
between
them.
METHODS
Erythrocytes
obtained
from
normal
male
and
female
volun-
teers
(19-34
yr
of
age)
were
washed
three
times
in
isos-
motic
choline
chloride
(pH
=
7.4).
Ouabain
binding
was
determined
as
described
previously
(1).
Erythrocytes
were
added
to
incubation
solutions
(pre-
warmed
to
37°C
unless
otherwise
specified)
containing
['H]
ouabain.
The
hematocrit
of
the
incubation
mixture
was
5-10%.
After
mixing
thoroughly,
triplicate
100-,ul
samples
were
taken
at
appropriate
times,
placed
in
poly-
ethylene
micro
test
tubes
(Beckman
Instruments,
Inc.,
Fullerton,
Calif.)
and
washed
five
times
with
300
,ul
of
isosmotic
choline
chloride
by
alternate
centrifugation
and
resuspension.
Centrifugation
was
performed
using
a
Micro-
fuge
(Beckman
Instruments,
Inc.)
at
10,000
g
for
15
s.
After
the
final
wash,
each
sample
was
treated
with
100
,u1
The
Journal
of
Clinical
Investigation
Volume
52
August
1973a
1845-1851
1845

of
10%o
perchloric
acid,
agitated,
and
centrifuged
for
30
s.
The
tube
and
its
contents
were
inverted
and
placed
in
a
vial
containing
20
ml
of
liquid
scintillation
solution.
When
the
vial
was
capped
and
shaken,
the
supernate
passed
from
the
sample
tube
into
the
scintillator
and
the
precipitate
remained
in
the
tip
of
the
sample
tube.
At
some
time
during
the
incubation,
triplicate
100-Al
samples
of
the
incubation
mixture
were
added
to
100
A.l
of
10%o
perchloric
acid,
agi-
tated,
centrifuged,
inverted,
and
placed
in
a
vial
containing
liquid
scintillation
solution.
The
volume
of
cells
counted
was
calculated
from
the
hemoglobin
concentration
and
hematocrit
determined
on
a
separate
tube
containing
the
incubation
solution
and
erythrocytes
at
an
hematocrit
of
approximately
25%o.
The
hematocrit
was
measured
using
a
Drummond
microhematocrit
centrifuge
(Drummond
Scien-
tific
Co.,
Broomall,
Pa.)
and
hemoglobin
concentration
was
measured
using
the
cyanmethemoglobin
method
(5).
The
standard
incubation
solution
had
the
following
com-
position
(millimolars):
NaCl,
150;
Tris
buffer
(pH
=
7.4),
10;
glucose,
11.1.
Whenever
the
cation
composition
of
the
incubation
solution
was
altered,
isosmotic
choline
was
used
as
a
replacement.
The
amount
of
ouabain
bound
was
calculated
from
the
counts
per
milliliter
of
cells
and
the
specific
activity
of
ouabain
in
the
incubation
medium.
Binding
of
radioactive
impurities
or
entrapment
of
[3H]ouabain
was
determined
from
the
number
of
counts
bound
in
the
presence
of
10,000-
fold
molar
excess
of
nonradioactive
ouabain.
The
concen-
tration
of
ouabain
in
the
[3H]ouabain
supplied
by
the
manu-
facturer
was
determined
as
described
previously
(1)
by
measuring
the
binding
of
radioactivity in
the
presence
of
constant
[3H]
ouabain
and
varying
concentrations
of
non-
radioactive
ouabain.
The
concentration
thus
determined
was
within
11%
of
the
value
calculated
from
the
concentration
given
by
the
commercial
supplier.
In
previous
studies
we
have
demonstrated
that
all
of
the
cell-associated
radio-
activity
which
is
detected
using
this
technique
represents
['H]ouabain
bound
to
the
erythrocyte
membrane
(1,
6).
Liquid
scintillation
counting
was
performed
using
20
ml
of
a
solution
composed
of
15
parts
toluene
(J.
T.
Baker
Chemical
Co.,
Phillipsburg,
N.
J.),
5
parts
Triton
X-100
(New
England
Nuclear,
Boston,
Mass.),
and
1
part
Liqui-
fluor
(New
England
Nuclear).
The
observed
counts
were
usually
such
that
their
standard
deviation
was
less
than
2%o.
Variation
in
quenching
was
monitored
by
using
the
ratio
of
counts
in
two
channels
produced
by
an
automatic
external
standard;
however,
since
in
all
cases
the
maximum
range
was
less
than
3.2%,
no
quench
correction
was
made.
To
explore
the
relation
between
ouabain
binding
and
cation
transport,
potassium
influx
was
measured
on
cells
which
had
been
incubated
with
or
without
['H]ouabain.
At
appropriate
times,
triplicate
samples
were
taken
for
deter-
mination
of
ouabain
binding
and
an
additional
sample
was
removed
and
washed
three
times
with
30
vol
of
cold
(4'C)
isosmotic
choline
chloride.
To
determine
potassium
influx,
these
washed
cells
were
added
to
incubation
solutions
(37'C)
containing
42K.
The
hematocrit
was
4%
or
less.
After
mixing
thoroughly,
duplicate
100-,l
samples
were
placed
in
polyethylene
micro
test
tubes
and
washed
four
times
with
300
,ul
of
cold,
isosmotic
sodium
chloride.
After
the
final
wash
each
sample
was
treated
with
100
Al
of
10%
perchloric
acid,
agitated,
centrifuged,
inverted,
and
placed
in
20
ml
of
liquid
scintillation
fluid
for
counting.
At
some
time
during
the
incubation,
triplicate
100-,ul
samples
of
the
incubation
mixture
(i.e.,
cells
plus
medium)
were
added
to
a
micro
test
tube
containing
100
,Al
10%
perchloric
acid,
agitated,
centrifuged,
inverted,
and
placed
in
liquid
scintillation
solution.
The
volume
of
cells
counted
was
cal-
culated
from
the
hemoglobin
concentration
in
the
incubation
mixture
and
the
previously
measured
hemoglobin
content
per
volume
of
cells.
Erythrocyte
sodium
and
potassium
concentrations
were
determined
as
described
previously
(7).
The
incubation
solution
used
to
determine
potassium
influx
had
the
following
composition
(millimolars):
NaCl,
150;
Tris
buffer
(pH
=
7.4),
10;
glucose,
11.1.
The
potas-
sium
concentration
varied
depending
on
the
experimental
conditions.
The
sodium
and
potassium
concentrations
of
the
incubation
solutions
were
also
determined
at
the
end
of
the
incubation
period.
Initially
the
uptake
of
'K
was
deter-
mined
at
0,
15,
30,
and
45
min;
however,
since
the
uptake
was
observed
to
be
constant
over
this
period,
potassium
influx
was
calculated
from
samples
taken
at
0
and
40
min
unless
otherwise
specified.
Potassium
influx
was
cal-
culated
using
the
method
described
by
Sachs
and
Welt
(2)
and
the
average
of
the
potassium
concentrations
in
the
incubation
solutions
at
0
and
40
min.
All
counts
were
cor-
rected
for
decay.
Sodium
and
potassium
concentrations
were
measured
with
an
Instrumentation
Laboratory
model
143
flame
photometer
(Instrumentation
Laboratory,
Inc.,
Lexington,
Mass.).
Digoxin
was
kindly
supplied
in
crystalline
form
by
Dr.
Stanley
T.
Bloomfield,
Burroughs
Wellcome
Co.,
Research
Triangle
Park,
N.
C.
All
other
reagents
were
of
the
highest
grade
of
purity
obtainable.
[3H]Ouabain
(Lot
no.
184-194,
sp
act
13
Ci/mmol)
was
obtained
from
New
England
Nuclear,
and
radiochemical
purity
was
greater
than
97%
by
the
supplier's
radiochromatographic
and
reverse
isotope
dilution
criteria.
42K
was
obtained
as
the
chloride
from
ICN
Corp.,
Chemical
&
Radioisotopes
Div.,
Irvine,
Calif.
RESULTS
We
have
previously
reported
that
altering
the
cation
composition
of
the
incubation
medium
changes
both
the
rate
at
which
ouabain
is
bound
to
the
erythrocyte
mem-
brane
and
the
amount
bound
at
the
steady
state
(1).
The
data
illustrated
in
Fig.
1
indicate
that
the
rate
at
which
bound
ouabain
is
lost
from
the
cell
membrane
is
not
de-
pendent
on
the
cation
composition
of
the
incubation
me-
dium.
Values
obtained
in
the
presence
of
sodium
(which
increases
ouabain
binding
relative
to
choline)
were
not
significantly
different
from
values
obtained
in
the
pres-
ence
of
potassium
(which
decreases
ouabain
binding
rela-
tive
to
choline).
There
was
a
good
correlation
between
the
loss
of
ouabain
molecules
from
the
membrane
and
the
recovery
in
ouabain-sensitive
potassium
influx
(Fig.
1,
insert).
Fig.
2
illustrates
ouabain
binding
and
potassium
in-
flux
in
erythrocytes
which
had
been
preincubated
for
3
h
with
various
concentrations
of
[3H]ouabain.
In
these
studies,
potassium
influx
was
determined
using
a
potas-
sium
concentration
which
was
sufficient
to
produce
nearly
maximal
values
for
ouabain-sensitive
potassium
influx
(2).
As
the
ouabain
concentration
in
the
incubation
me-
dium
was
increased,
there
was
a
curvilinear
increase
in
ouabain
binding
until
a
ouabain
concentration
was
reached
above
which
further
increases
produced
no
fur-
1846
J.
D.
Gardner
and
D.
R.
Kiino

1.00
0.80p
e
2
0
0
z
0
4
.0
0.601
0.40[
0.20
w
1.00
1.00
z
ZZ
~~~~0
W
-0.60
0.60
O
4CfU.
..
0.20
I
0.20
0
TIME
(h)
I
I
1
2
3
4
5
TIME
(h)
FIGURE
1
Effect
of
altering
the
cation
composition
of
the
incubation
medium
on
the
loss
of
bound
ouabain
from
the
erythrocyte
membrane.
Erythrocytes
were
preincubated
for
3
h
in
a
solution
containing
sodium
(150
mM)
and
['H]-
ouabain
(1.1
X
10'
M).
The
cells
were
then
washed
rapidly
four
times
with
iced
(4'C)
isosmotic
choline
chlo-
ride
and
resuspended
in
the
appropriate
incubation
solu-
tions
at
370C.
At
the
indicated
times,
the
amount
of
ouabain
bound
to
the
erythrocyte
membrane
was
determined
and
the
values
are
expressed
as
the
fraction
of
['H]ouabain
bound
to
the
cells
at
the
beginning
of
the
incubation
period.
The
cation
composition
of
the
incubation
solutions
studied
were
sodium,
150
mM
(closed
circles)
;
choline,
150
mM
(open
circles);
potassium,
20
mM,
choline
130
mM
(open
boxes).
Each
point
represents
the
mean
of
three
experiments.
Insert.
Comparison
of
the
time
course
for
the
loss
of
ouabain
from
the
erythrocyte
membrane
and
that
for
the
recovery
of
ouabain-sensitive
potassium
influx.
Erythrocytes
were
preincubated
for
3
h
in
a
solution
con-
taining
sodium
(150
mM)
and
['H]ouabain
(1.21
X
10'
M).
The
cells
were
then
washed
rapidly
four
times
with
iced
(4°C)
isosmotic
choline
chloride
and
resuspended
in
incubation
solution
(sodium,
150
mM)
at
37°C.
At
the
indicated
times
the
amount
of
ouabain
bound
to
the
erythro-
cyte
membrane
was
determined
and
the
values
we
expressed
as
the
fraction
of
['H]ouabain
bound
to
the
cell
membrane
at
the
beginning
of
the
incubation
period.
At
the
indicated
times
samples
were
also
taken,
added
to
tubes
containing
"K
and
potassium
influx
determined
over
a
15
min
period.
Values
for
potassium
influx
are
expressed
as the
fraction
of
glycoside-sensitive
potassium
influx
determined
on
cells
which
had
been
preincubated
for
3
h
with
and
without
10'
M
ouabain.
The
potassium
concentration
used
to
deter-
mine
potassium
influx
was
10.3
mM.
Each
point
represents
the
mean
of
two
experiments.
ther
increase
in
ouabain
binding.
A
similar
relation
was
observed
between
ouabain
concentration
and
inhibi-
tion
of
potassium
influx.
That
is,
as
the
ouabain
con-
centration
was
increased
there
was
a
curvilinear
de-
crease
in
potassium
influx
until
a
ouabain
concentration
was
reached
above
which
there
was
no
further
decrease
in
potassium
influx.
There
was
good
agreement
between
the
ouabain
concentration
at
which
ouabain
binding
was
half-maximal
and
that
at
which
inhibition
of
potassium
influx
was
half-maximal.
Fig.
3
illustrates
the
results
of
an
experiment
similar
to
that
illustrated
in
Fig.
2
except
that
potassium
in-
flux
was
measured
at
a
low
extracellular
potassium
con-
centration
(i.e.
one
such
that
ouabain-sensitive
potassium
influx
was
15%
of
that
observed
for
the
experiments
illustrated
in
Fig.
2).
Although
the values
for
potassium
influx
in
these
experiments
were
appreciably
lower
than
those
illustrated
in
Fig.
2,
a
similar
relation
between
ouabain
concentration
and
potassium
influx
was
observed
and
there
was
good
agreement
between
the
ouabain
con-
centration
at
which
ouabain
binding
was
half-maximal
and
that
at
which
inhibition
of
potassium
influx
was
half-
maximal.
Furthermore,
even
though
the
influx
values
il-
lustrated
in
Fig.
2
were
significantly
higher
than
those
illustrated
in
Fig.
3,
there
was
close
agreement
between
the
two
experiments
for
the
ouabain
concentration
at
which
inhibition
of
potassium
influx
was
half-maximal.
To
further
explore
this
relation
between
ouabain
bind-
ing
and
potassium
influx,
ouabain
binding
was
measured
at
five
different
ouabain
concentrations
and
then
potas-
sium
influx
was
measured
on
cells
from
each
different
ouabain
concentration
using
five
different
external
po-
tassium
concentrations.
At
each
potassium
concentra-
tion
studied
as
ouabain
binding
increased,
there
was
a
progressive
constant
decrease
in
the
values
for
potassium
2.4
cr
=L
D2
0
0
^
-
'a
_
-1
U.
s
X.
0I
2
4
6
8
OUABAIN
CONCENTRATION,
(x108M)
24
20
16
0
12
,,
E
z
0
4
40
10
FIGURE
2
Ouabain
binding
and
potassium
influx
as
a
func-
tion
of
the
ouabain
concentration
in
the
incubation
medium.
Erythrocytes
were
incubated
for
3
h
(37'C)
in
an
incuba-
tion
solution
containing
sodium,
150
mM
and
various
con-
centrations
of
['H]ouabain.
Samples
were
taken
for
de-
termination
of
ouabain
binding
and
then
"K
was
added
to
the
incubation
medium
and
potassium
influx
was
determined.
The
potassium
concentration
in
the
influx
medium
was
8.3
mM.
Each
point
represents
the
mean
of
two
experiments.
Ouabain
Binding
and
Cation
Transport
1847
2.0
1.0

0.32,
_
U.
C-0
-i
0.24
=
020
76
E
E-
°2
0
16
z
0
12-
2
en
01
-
008
0
a.
7
-w
a
OUABAIN
CONCENTRATION
(x
I0-8M)
FIGURE
3
Ouabain
binding
and
potassium
influx
as
a
func-
tion
of
the
ouabain
concentration
in
the
incubation
me-
dium.
The
procedure
was
identical
to
that
given
in
the
legend
to
Fig.
3
except
that
the
potassium
concentration
used
to
measure
potassium
influx
was
0.37
mM.
Each
point
represents
the
mean
of
two
experiments.
influx
and
the
magnitude
of
this
decrease
tended
to
be
less
at
lower
potassium
concentrations
(Fig.
4).
A
more
meaningful
presentation
of
the
data
in
Fig.
4
is
illustrated
by
the
insert
where
the
data
for
potassium
influx
are
plotted
as
the
fraction
of
glycoside-sensitive
potassium
influx
observed
at
a
given
potassium
concen-
tration.
This
mode
of
presentation
permits
one
to
com-
pare
directly
the
fraction
of
ouabain-binding
sites
which
are
occupied
by
ouabain
with
the
fraction
of
ouabain-
sensitive
potassium
influx
which
is
inhibited.
As
is
indi-
cated,
for
a
given
value
of
ouabain
binding,
values
for
the
fraction
of
ouabain-sensitive
potassium
influx
were
not
significantly
different.
Furthermore,
these
data
were
best
described
by
a
single
straight
line
having
intercepts
on
the
x-
and
y-axis
of
1.0.
That
is,
occupation
of
a
given
fraction
of
glycoside-binding
sites
by
ouabain
in-
hibits
an
equal
fraction
of
ouabain-sensitive
potassium
influx
and
this
relation
is
independent
of
the
potassium
concentration
in
the
incubation
solution
and
of
the actual
magnitude
of
glycoside-sensitive
potassium
influx.
Since
each
of
the
studies
illustrated
in
Figs.
2-4
involved
determination
of
ouabain
binding
in
incubation
solutions
containing
150
mM
sodium,
we
felt
it
was
important
to
explore
further
the
relation
between
ouabain
binding
and
potassium
influx
by
measuring
ouabain
binding
in
solutions
having
different
cation
compositions.
Fig.
5
illustrates
values
for
ouabain
binding
determined
on
cells
which
had
been
incubated
in
a
solution
containing
cho-
line
(150
mM)
or
choline
(130
mM)
and
potassium
(20
mM)
and
various
concentrations
of
[3H]ouabain.
At
the
end
of
the
incubation,
samples
were
taken
for
determination
of
ouabain
binding.
The
remaining
cells
were
washed
three
times
and
resuspended
in
a
solution
containing
150
mM
sodium.
'K
was
added
to
the
incu-
bation
mixture
and
potassium
influx
was
determined.
Under
these
conditions
the
values
obtained
for
ouabain
binding
and
for
the
inhibition
of
potassium
influx
were
significantly
lower
than
those
obtained
when
the
incu-
bation
medium
contained
sodium,
150
mM
(see
Figs.
2
and
3).
However,
as
was
observed
when
ouabain
bind-
ing
was
determined
in
the
presence
of
sodium,
150
mM,
increasing
the
ouabain
concentration
produced
a
pro-
gressive
increase
in
ouabain
binding
and
a
corresponding
decrease
in
potassium
influx.
To
see
if
altering
the
ca-
tion
composition
of
the
incubation
solution
altered
the
relation
between
ouabain
binding
and
inhibition
of
potas-
sium
influx,
the
data
in
Fig.
5
were
plotted
in
the
form
of
fraction
of
ouabain-sensitive
potassium
influx
vs.
the
fraction
of
the
total
ouabain-binding
sites
occupied
(Fig.
1.00
wx
(02
-
020
~2.0
0
0.40
0.80
QUASAIN
BOUND
(Fraction)
1.6
E
E
X
1.2
U.
2
63
~~~~~~~~~~~~~~12.0
C0
0.4~~~~~~~~~~~~.
3.6
1.8
I
,~~~~~~~0.4
0
0.20
0.40
0.60
0.80
1.00
OUABAIN
BOUND
(Fraction)
FIGURE
4
Potassium
influx
measured
at
different
potassium
concentrations
as
a
function
of
ouabain
bound
to
the
eryth-
rocyte
membrane.
Erythrocytes
were
incubated for
3
h
(370C)
in
solutions
containing
sodium,
150
mM
and
various
concentrations
of
[8H]
ouabain.
Ouabain
binding
was
de-
termined
and
thqn
potassium
influx
was
measured
on
cells
which
had
been
incubated
at
each
different
ouabain
con-
centration
using
five
different
potassium
concentrations.
The
potassium
concentrations
(millimolars)
used
to
measure
potassium
influx
are
indicated
at
the
lower
righthand
por-
tion
of
the
figure.
Ouabain
binding
is
given
as
the
fraction
of
the
maximal
amount
of
ouabain
which
could
be
bound.
The
values
represent
the
mean
of
three
experiments.
Insert
For
each
different
potassium
concentration
studied,
ouabain-
sensitive
potassium
influx
was
calculated
as
the
fraction
of
the
total
ouabain-sensitive
potassium
influx
which
was
present
at
that
particular
potassium
concentration
and
was
plotted
against
the
fraction
of
the
maximal
amount
of
ouabain
which
could
be
bound.
1848
J.
D.
Gardner
and
D.
R.
Kiino

0.70
X
0.50
so
E
0
0.40
x-
Z
0.30
i-
0.20
2
0.10[
I-
20
16
2
.
a
-i
z
m
a
co
0
4
8
12
16
20
OUABAIN
cONCENTRATION
(x
117M)
FIGURE
5
Ouabain
binding
and
potassium
influx
as
a
function
of
ouabain
concentration
for
cells
incubated
with
choline
or
potassium.
Erythrocytes
were
incubated
for 3
h
(370C)
in
solutions
containing
different
concentrations
of
['H]ouabain
and
choline,
150
mM
(circles)
or
choline,
130
mM
plus
potassium,
20
mM
(boxes).
Samples
were
taken
for
determination
of
ouabain
binding
and
the
cells
were
washed
rapidly
four
times
with
iced
(40C)
isosmotic
choline
chloride
and
resuspended
in
a
solution
containing
sodium,
150
mM
at
370C.
Potassium
influx
was
then
determined
using
"K
and
a
potassium
concentration
of
1.30
mM.
The
values
represent
the
mean
of
two
experiments.
6).
The
observed
linear
relation
indicates
that
regard-
less
of
the
cation
composition
of
the
incubation
solution
used
to
determine
ouabain
binding
or
of
the
magnitude
of
ouabain-sensitive
potassium
influx,
when
a
given
frac-
tion
of
the
ouabain-binding
sites
is
occupied,
there
is
a
corresponding
fractional
inhibition
of
ouabain-sensitive
potassum
influx.
DISCUSSION
The
erythrocyte
membrane
contains
a
finite
number
of
glycoside-binding
sites
whose
affinity
for
cardioactive
glycosides
is
dependent
on
the
cation
composition
of
the
incubation
medium
(1).
It
is
also
known
that
a
portion
of
potassium
influx
across
the
erythrocyte
membrane
can
be
inhibited
by
cardioactive
glycosides
and
that
this
gly-
coside-sensitive
potassium
influx
is
a
saturable
function
of
the
extracellular
potassium
concentration
and
can
be
inhibited
competitively
by
other
monovalent
cations
(2).
The
present
studies
were
designed
to
explore
the
func-
tional
characteristics
of
the
mechanism
by
which
cardio-
active
glycosides,
once
bound
to
the
erythrocyte
mem-
brane,
inhibit
erythrocyte
potassium
influx.
The
rate
at
which
bound
ouabain
dissociates
from
the
cell
membrane
is
not
altered
by
complete
replacement
of
choline
with
sodium
or
by
partial
replacement
with
po-
tassium
(Fig.
1).
These
findings
indicate
that
the
previ-
ous
observations
that
sodium
increases
and
that
potas-
sium
decreases
ouabain
binding
(1)
can
be
attributed
to
the
effects
of
these
cations
on
the
rate
of
association
of
ouabain
with
the
erythrocyte
membrane.
The
good
cor-
relation
observed
between
the
decrease
in
cell-associated
radioactivity
and
the
rise in
ouabain-inhibitable
potas-
sium
influx
(Fig.
1,
insert)
indicates
that
this
decrease
in
radioactivity
represents
the
loss
of
pharmacologically
active
ouabain
molecules
from
the
cell
membrane
and
ex-
cludes
the
possibility
that
the
decrease
in
radioactivity
represents
the
loss
of
a
tritiated
contaminant.
Further-
more,
the
relatively
slow
rate
at
which
ouabain
was
lost
from
the
membrane
indicates
that
during
the
incubation
period
used
to
measure
potassium
influx
the
loss
of
ouabain
from
the
erythrocyte
membrane
was
negligible.
In
agreement
with
a
previous
study
from
this
labora-
tory
(1)
our
value
for
maximal
ouabain
binding
(18.84
1.7
pmol/ml
cells)
indicates
that
there
are
approximately
1,100
glycoside-binding
sites
per
erythrocyte.
This
value
is
significantly
higher
than
those
reported
previously
by
others
(8-10).
Part
of
this
discrepancy
appears
to
be
at-
1.00
U.
z
20.80
0
2
U)
I<_
2
0.60
Z
0.40
z
\
90.20
0
0.20
0.40
0.60
0.80
.00
OUABAIN
BOUND
(Fraction)
FIGURE
6
Ouabain-sensitive
potassium
influx
as
a
function
of
ouabain
bound
for
cells
incubated
in
solutions
with
dif-
ferent
cation
compositions.
The
values
for
choline
(closed
circles)
and
potassium
(open
circles)
were
calculated
from
the
values
given
in
Fig.
5.
The
values
for
sodium
(closed
boxes)
were
obtained
using
the
same
procedure
as
outlined
in
the
legend
to
Fig.
5
except
that
erythrocytes
were
pre-
incubated
for
3
h
(370C)
in
a
solution
containing
sodium,
150
mM
and
different
concentrations
of
['H]
ouabain.
The
values
for
ouabain
binding
are
expressed
as
the
fraction
of
the
value
for
maximal
ouabain
binding
determined
by
in-
cubating
the
cells
in
the
appropriate
solutions
with
a
con-
centration
of
['H]
ouabain
sufficient
to
produce
maximal
ouabain
binding.
Ouabain-sensitive
potassium
influx
was
calculated
as
the
fraction
of
the
total
ouabain-sensitive
po-
tassium
influx
which
was
present
in
cells
which
had
been
incubated
in
each
of
the.three
different
incubation
solutions.
Ouabain
Binding
and
Cation
Transport
1849
'D
-
19
II
.8
4
I-

tributable
to
differences
in
the
method
for
extracting
bound
radioactivity
from
the
cells,
in
the
composition
of
the
liquid
scintillation
mixture
and
the
procedure
used
to
correct
for
variable
counting
efficiency
(1).
Another
po-
tential
source
of
this
discrepancy
may
be
the
different
lots
of
['H]ouabain
used.
The
specific
activity
of
the
['H]ouabain
used
in
the
present
experiments
was
3-25
times
greater
than
that
used
by
others
(8-10).
Dunham
and
Hoffman
(11)
also
noted
that
they
obtained
sig-
nificantly
different
values
for
ouabain
binding
to
sheep
erythrocytes
when
they
used
a
different
lot
of
['H].oua-
bain
from
the
same
commercial
supplier
and
speculated
that
variation
among
different
lots
of
[8H]ouabain
might
account
for
the
differences
in
values
for
ouabain
binding
to
sheep
erythrocytes
obtained
by
different
laboratories.
When
the
potassium
concentration
in
the
incubation
solution
is
0.4
mM,
glycoside-sensitive
potassium
influx
is
approximately
15%
of
what
it
is
when
the
potassium
concentration
in
the
incubation
solution
is
12
mM
(Fig.
4).
One
question
which
the
present
studies
were
designed
to
answer
was
whether
more,
less,
or
the
same
amount
of
bound
ouabain
is
required
to
produce
50%
inhibition
of
potassium
influx
measured
at
a
potassium
concentration
of
0.4
mM
compared
with
the
amount
required
to
pro-
duce
50%
inhibition
of
potassium
influx
measured
at
a
potassium
concentration
of
12
mM.
It
seemed
that
when
glycoside-sensitive
potassium
influx
was
submaximal,
less
bound
ouabain
would
be
required
to
produce
50%
inhibition
than
when
potassium
influx
was
maximal.
However,
as
is
illustrated
by
Fig.
4
the
same amount
of
bound
ouabain
is
required
to
produce
50%
inhibition
of
potassium
influx
measured
at
0.4
mM
potassium
as
is
required
when
potassium
influx
is
measured
at
12
mM
potassium.
The
relation
between
ouabain
binding
and
in-
hibition
of
potassium
influx
is
that
irrespective
of
the
potassium
concentration
used
to
measure
potassium
influx,
the
fraction
of
the
glycoside-inhibitable
potassium
influx
which
was
inhibited
was
equal
to
the
fraction
of
the
gly-
coside-binding
sites
which
were
occupied
by
ouabain.
We
have
previously
demonstrated
that
in
terms
of
its
effect
on
ouabain
binding,
potassium
acts
at
a
site
which
also
has
an
affinity
for
sodium
and
which
is
functionally
distinct
from
the
site
to
which
cardiac
glycosides
bind
(1).
When
this
cation
site
is
occupied
by
potassium,
the
affinity
of
the
glycoside-binding
site
for
glycosides
is
decreased.
We
have
also
observed
that
there
is
a
potas-
sium
concentration
above
which
no
further
decrease
in
ouabain
binding
was
observed
and
the
inhibition
of
ouabain
binding
produced
by
a
given
concentration
of
potassium
could
be
overcome
by
increasing
the
concen-
tration
of
ouabain
in
the
incubation
medium.
Hoffman
and
co-workers
concluded
that
although
ouabain
binding
could
be
abolished
by
raising
the
external
potassium
concentration
to
30
mM
(8,
9),
cardiac
glycosides
were
noncompetitive
inhibitors
of
glycoside-sensitive
cation
transport
because
cesium
could
substitute
for
potassium
in
activating
sodium
outflux
but
cesium
was
not
able
to
replace
potassium
in
preventing
or
altering
the
action
of
cardiac
glycosides
on
cation
transport
(4).
We
have
specified
previously
that
the
technique
used
by
Hoffman
and
Ingram
was
probably
not
sufficiently
sensitive
to
de-
tect
ouabain
binding
in
the
presence
of
30
mM
potassium
(1).
Glynn
concluded
that
cardiac
glycosides
competi-
tively
inhibited
the
interaction of
potassium
ions
with
the
glycoside-sensitive
transport
mechanism
(3).
This
conclusion
was
based
on
the
observation
that
the
inhi-
bition
of
potassium
influx
by
cardiac
glycosides
was
re-
duced
as
the
concentration
of
potassium
in
the
incuba-
tion
medium
was
increased.
These
studies,
however,
did
not
permit
one
to
distinguish
between
effects
of
potas-
sium
on
glycoside
binding
and
effects
of
potassium
on
potassium
influx.
The
data
in
Fig.
4
exclude
the
possibility
that
binding
of
ouabain
molecules
to
the
glycoside-binding
site
reduces
but
does
not
abolish
the
apparent
affinity
of
potassium
ions
for
the
potassium
transport
mechanism.
If
bound
glycoside
molecules
reduced
but
did
not
abolish
the
affin-
ity
of
the
potassium
transport
sites
for
potassium,
at
a
given
value
of
ouabain
binding,
one
would
expect
to
find
that
the
fraction
of
glycoside-inhibitable
potassium
trans-
port
decreased
with
increasing
concentrations
of
external
potassium.
Furthermore,
this
decrease
should
be
more
readily
observed
at
lower
values
for
ouabain
binding
than
at
higher
values.
It
might
be
argued
that
we
did
not
use
a
sufficiently
high
concentration
of
potassium
to
detect
such
an
effect;
however,
even
at
the
lowest
values
of
ouabain
binding
studied
raising
the
external
potassium
concentration
to
30
mM
did
not
alter
the
fractional
in-
hibition
of
glycoside-sensitive
potassium
influx.
Instead,
our
findings
indicate
that
cardiac
glycosides
bind
to
a
site
which
is
functionally
distinct
from
the
po-
tassium
transport
site'
and
when
the
glycoside-binding
sites
are
occupied
by
ouabain,
a
proportional
number
of
potassium
transport
sites
are
inhibited.
That
is,
ouabain
acts
to
abolish
the
affinity
of
the
potassium
transport
sites
for
potassium
and
the
fraction
of
potassium
trans-
port
sites
so
altered
is
equal
to
the
fraction
of
glycoside-
binding
sites
occupied
by
ouabain.
When
ouabain
mole-
cules
are
bound
to
the
erythrocyte
membrane,
the
potas-
sium
transport
sites
can
be
viewed
as
existing
in
one
of
three
different
functional
configurations.
A
given
number
of
potassium
transport
sites
are
unable
to
combine
with
potassium
ions
as
a
result
of
their
being
inhibited
by
1
The
use
of
the
term
"transport
site"
should
not
be
taken
to
indicate
that
such
an
entity
actually
exists.
This
term
is
used
only
to
facilitate
our
expression
of
certain
salient
functional
characteristics
of
erythrocyte
cation
transport.
1850
J.
D.
Gardner
and
D.
R.
Kiino

ouabain
molecules
which
are
bound
to
the erythrocyte
membrane.
The
potassium
transport
sites
which
are
not
inhibited
may
be
either
active
or
inactive
(in
terms
of
being
involved
in
the
translocation
of
potassium
ions
across
the
erythrocyte
membrane)
depending
on
the
po-
tassium
concentration
in
the
incubation
medium.
Further-
more,
for
the
uninhibited
sites,
the
ratio
of
active
sites
to
inactive
sites
is
independent
of
the
number
of
glycoside
molecules
bound
to
the
erythrocyte
membrane.
A
second
question
which
the
present
studies
were
de-
signed
to
answer
was
whether
altering
the
cation
com-
position
of
the
incubation
medium
altered
the
relation
between
ouabain
binding
and
inhibition
of
potassium
influx.
Figs.
5
and
6
illustrate
that
although
varying
the
cation
composition
of
the
incubation
solution
alters
the
affinity
of
the
glycoside-binding
sites
for
ouabain
and
alters
the
magnitude
of
glycoside-sensitive
potassium
influx,
these
variations
did
not
alter
the
relation
between
the
fraction
of
glycoside-binding
sites
occupied
by
oua-
bain
and
the
fraction
of
glycoside-sensitive
potassium
in-
flux
which
was
inhibited.
These
observations
indicate
that
irrespective
of
the
affinity
of
the
erythrocyte
mem-
brane
for
ouabain
and
irrespective
of
the
magnitude
of
glycoside-sensitive
potassium
influx,
occupation
of
a
given
fraction
of
the
glycoside-binding
sites
by
ouabain
results
in
the
inhibition
of
an
equal
fraction
of
the
oua-
bain-sensitive
potassium
transport
sites.
The
present
studies
do
not
indicate
whether
or
not
the
site
at
which
potassium
ions
act
to
alter
the
affinity
of
the
glycoside-binding
site
for
glycoside
molecules
also
functions
as
a
potassium
transport
site.
The
observations
that
potassium
decreases
but
does
not
abolish
the
affinity
of
the
glycoside-binding
site
for
ouabain
but
that
gly-
coside
molecules
when
bound
to
their
binding
sites
abolish
the
affinity
of
the
potassium
transport
sites
for
potassium
might
appear
to
indicate
that
the
site
at
which
potassium
ions
act
to
alter
the
glycoside
site
must
be
functionally
distinct
from
that
site
at
which
potassium
acts
to
be
transported
across
the
erythrocyte
membrane.
However,
as
we
have
demonstrated
previously
(1),
the
observed
effects
of
potassium
ions
on
glycoside
binding
are
compatible
with
the
possibility
that
when
the
glyco-
side-binding
site
is
occupied
by
ouabain
the
affinity
of
the
monovalent
cation
site
for
potassium
is
abolished.
In
other
words,
potassium
ions
can
combine
with
the
cation
site
only
when
the
glycoside-binding
site
is
not
occupied
by
glycoside
molecules.
If
this
is,
in
fact,
the
situation
and
if
the
cation
site
is
also
a
potassium
trans-
port
site,
this
effect
of
cardiac
glycosides
could
account
for
their
ability
to
inhibit
potassium
influx.
Sachs
and
Welt
(2)
found
that
the
component
of
po-
tassium
influx
which
was
abolished
in
cells
which
had
been
depleted
of
energy
stores
and
incubated
with
10i
M
strophanthidin
required
two
potassium
ions
to
be
present
at
some
site
or
sites
in
the
transport
mechanism
before
transport
occurred.
These
authors'
data
also
indicate
that
the
two
sites
have
different
affinities
for
potassium.
If
the
site
at
which
potassium
acts
to
alter
the
affinity
of
the
glycoside
site
for
ouabain
also
functions
as
one
of
the
two
potassium
transport
sites
described
by
Sachs
and
Welt,
our
previous
studies
would
indicate
that
it
is
the
site
with
the
higher
affinity
for
potassium
(1).
From
the
data
given
by
Sachs
and
Welt
we
have
calculated
that
the
value
for
the
dissociation
constant
describing
the
interaction
between
potassium
ions
and
the
higher
affinity
transport
site
is
0.37
mM.
The
value
which
we
have
previ-
ously
reported
for
the
dissociation
constant
describing
the
interaction
between
potassium
ions
and
the
site
at
which
sodium
and
potassium
act
to
alter
ouabain
binding
is
0.28
(+
0.17)
mM
(mean
[+SD]).
However,
the
fact
that
these
two
values
are
not
significantly
different
by
no
means
proves
that
they
reflect
the
action
of
potas-
sium
at
the
same
site
and
additional
studies
are
necessary
before
this
question
can
be
resolved.
ACKNOWLEDGMENTS
We
thank
Blanche
Fors
and
Agnes
Sady
for
typing
the
manuscript.
REFERENCES
1.
Gardner,
J.
D.,
and
T.
P.
Conlon.
1972.
The
effects
of
sodium
and
potassium
on
ouabain
binding
by
human
erythrocytes.
J.
Gen.
Physiol.
60:
609.
2.
Sachs,
J.
R.,
and
L.
G.
Welt.
1967.
The
concentration
dependence
of
active
potassium
transport
in
the
human
red
blood
cell.
J.
Clin.
Invest.
46:
65.
3.
Glynn,
I.
M.
1964.
The
action
of
cardiac
glycosides
on
ion
movements.
Pharmacol.
Rev.
16:
381.
4.
Hoffman,
J.
F.
1966.
The
red
cell
membrane
and
the
transport
of
sodium
and
potassium.
Am.
J.
Med.
41:
666.
5.
Davidsohn,
I.,
and
B.
B.
Wells.
1962.
Clinical
Diagnosis
by
Laboratory
Methods.
W.
B.
Saunders
Co.,
Philadel-
phia.
73.
6.
Gardner,
J.
D.,
D.
R.
Kiino,
T.
J.
Swartz,
and
V.
P.
Butler,
Jr.
1973.
Effects
of
digoxin-specific
antibodies
on
the
accumulation
and
binding
of
digoxin
by
human
erythrocytes.
J.
Clin.
Invest.
52:
0000.
7.
Gardner,
J.
D.,
A.
Lapey,
A.
P.
Simopoulos,
and
E.
L.
Bravo.
1971.
Abnormal
membrane
sodium
transport
in
Liddle's
syndrome.
J.
Clin.
Invest.
50:
2253.
8.
Hoffman,
J.
F.,
and
C.
J.
Ingram.
1968.
Cation
trans-
port
and
the
binding
of
T-ouabain
to
intact
human
red
blood
cells.
Proceedings
of
the
1st
International
Sym-
posium
on
Metabolism
and
Membrane
Permeability
of
Erythrocytes
and
Thrombocytes,
Vienna.
420.
9.
Hoffman,
J.
F.
1969.
The
interaction
between
tritiated
ouabain
and
the
Na-K
pump
in
red
blood
cells.
J.
Gen.
Physiol.
54:
343s.
10.
Baker,
P.
F.,
and
J.
S.
Willis.
1972.
Binding
of
the
cardiac
glycoside
ouabain
to
intact
cells.
J.
Physiol.
(Lond.).
224:
441.
11.
Dunham,
P.
B.,
and
J.
F.
Hoffman.
1971.
Active
cation
transport
and
ouabain
binding
in
high
potassium
and
low
potassium
red
blood
cells
of
sheep.
J.
Gen.
Physiol.
58:
94.
Ouabain
Binding
and
Cation
Transport
1851