|
|
|
|||||||
|
|||||||||
(Received for publication, July 7, 1994; and in revised form, November 23, 1994) From the
Effects of lipophilic ions, tetraphenylphosphonium
(TPP Na Phosphoenzyme interactions with extracellular cations were evaluated through
K Effects of lipophilic
ions on the transition between phosphoenzymes were significantly
diminished when Na Lipophilic ions affected Na-ATPase, Na,K-ATPase, and p-nitrophenylphosphatase activities in accordance with their
effects on the partial reactions. Effects of lipophilic ions appear
to be due to their charge indicating that Na Lipophilic ions like tetraphenylphosphonium
(TPP In the
present work we applied biochemical techniques supported by
measurements of conformational changes by eosin fluorescence to
describe which partial reactions in the Na,K-ATPase reaction cycle are
affected by lipophilic ions. The studies were performed with broken
membrane Na,K-ATPase preparations; thus, the lipophilic ions had access
to both sides of the membrane and their distribution in the membrane
was not modified by a trans-membrane potential. To ease the
understanding of the experimental design, the following reaction scheme
is used as a frame of reference:
In this very simplified scheme, E The effect of
lipophilic ions on the following partial reactions were studied:
transitions between E
The
Na,K-ATPase activity of bovine brain, pig kidney, and shark rectal
gland enzymes, measured at 37 °C and under standard conditions
(Ottolenghi, 1975, but without bovine serum albumin) were 4-5,
27, and 25 units The protein amount was determined according to Lowry et al. (1951), as described by Jensen and Ottolenghi (1983a),
using bovine serum albumin as standard.
The fluorescence response of 0.5 µM eosin to
binding of K TPPCl and NaTPB dissolved in dimethyl sulfoxide were added to
final concentrations of 300 and 50 µM, respectively
(dimethyl sulfoxide
The description of the results follows the reaction : binding of Na
In our
experiments, specific eosin binding was induced by buffer (10 mM HEPES + 10 mM MES + 10 mM EDTA, pH 7.5
at 20 °C). The subsequent addition of 4 mM Na In the absence of Na
Figure 1:
The effect of
K
Thus, we observed the same
contradiction between affinities estimated from equilibrium and
transient experiments as described previously by Karlish et
al.(1978) and Glynn and Karlish(1982), who proposed the following
scheme to explain this discrepancy:
They assumed that a low affinity K In our
experiments, the rate coefficient of the specific eosin fluorescence
decrease in response to K Since K When 4 mM Na
Figure 2:
Equilibrium fluorescence change as
function of K
In this model, Na The experiments on Na
When the ionic strength was kept constant with NMGCl
or TrisCl, no effect of lipophilic ions on the formation of EP
was observed (Fig. 3). The same result was obtained when ionic
strength of the medium was increased by addition of 100 mM NMGCl or 100 mM TrisCl (not shown). Tris
Figure 3:
Effect of TPP
Figure 4:
Effect of lipophilic ions on Na-dependent
dephosphorylation. Dependence of dephosphorylation rate constant on
[NaCl]. Bovine brain Na,K-ATPase was phosphorylated for 60 s
at 0 °C as described under ``Materials and Methods'' at
varying [NaCl] in the absence of lipophilic ions (circles), with 300 µM TPP
In the absence of
lipophilic ions, the dephosphorylation was stimulated by NaCl in the
concentration range from 10 to 150 mM, similar to previous
data by Hara and Nakao(1981) and Nørby et al.(1983).
The stimulation appears to be due to a stimulation of dephosphorylation
of the K-sensitive EP. It has been shown previously that the
K-insensitive EP dephosphorylates directly (Nørby at al., 1983), and its transition to the K-sensitive EP is inhibited by NaCl (Klodos et al., 1994).
However, because of a low relative amount of
K
Figure 9:
TPP
Figure 8:
[KCl] dependence of K-stimulated
dephosphorylation of phosphoenzyme: Effect of TPP
The
addition of ADP in the chase produced rapid decay of the ADP-sensitive EP and exposed the slowly decaying ADP-insensitive EP (Fig. 5A). The biphasic dephosphorylation was analyzed
in the following scheme
Figure 5:
Effect of lipophilic ions on ADP-dependent
dephosphorylation. The experiments were performed with bovine brain
Na,K-ATPase as in Fig. 4at varying [NaCl] in the
absence of lipophilic ions (circles) and in the presence of
300 µM TPP
where EP In the absence of lipophilic
ions, an increase in [NaCl] resulted in: 1) an increase in
the steady state amount of ADP-sensitive EP (Fig. 5, A and B), and 2) an increase in the rate of the slow
decay (Fig. 5C). Both observations are in agreement
with previously published data (Hara and Nakao, 1981; Nørby et al., 1983; Klodos and Nørby, 1987). TPP Similar effects of lipophilic ions
were observed in experiments with Na,K-ATPase from pig kidney (Fig. 6) and shark rectal gland (not shown). Although the steady
state proportion of ADP-sensitive EP measured, in the presence
of 100 mM NaCl in the medium, was lower with these enzymes
than with brain Na, K-ATPase (compare Fig. 5and Fig. 6at 100 mM NaCl, cf. also Klodos &
Nørby(1987) and Klodos et al.(1994)), qualitative
effects of lipophilic ions on both the steady state level of
ADP-sensitive EP and the slope of the slow phase were the same
as with the brain Na,K-ATPase.
Figure 6:
ADP-stimulated dephosphorylation of pig
kidney Na,K-ATPase: effect of lipophilic ions. The experiment was the
same as in Fig. 5A except the source of enzyme. circles, NaCl alone; squares, +300 µM TPP
100 µM TPB
Figure 7:
Fraction of ADP-sensitive phosphoenzyme as
a function of [TPP
The phosphoenzyme formed in the
presence of 300 µM TPP The dephosphorylation
results, similar to those in Fig. 8, were evaluated according to
the following scheme
Again, EP Fig. 9shows the relative amount of the K-insensitive EP (Fig. 9A) and the apparent rate coefficient
of the slow (Fig. 9B) and the rapid (Fig. 9C) phases as a function of [NaCl]
without and with 300 µM TPP
Figure 10:
Effect of various salts and lipophilic
ions on ADP-dependent dephosphorylation. The experiments with bovine
brain enzyme were performed as in Fig. 5in the presence of 10
mM NaCl alone or 10 mM NaCl with either 90 mM NMGCl, 90 mM TrisCl, or 90 mM NaNO
As seen from Fig. 10and Fig. 11, the replacement of Na
Figure 11:
Effect of various salts and lipophilic
ions on K
None of
the substitutions affected TPP
Figure 12:
Time course of phosphorylation at three
different [NaCl]. The experiments were performed as described
under ``Materials and Methods'' with bovine brain Na,K-ATPase
and in the presence of 1 mM (left panel), 3 mM (middle panel), or 10 mM (right panel)
NaCl alone (circles); + 300 µM TPP
In accordance with Fig. 3, the
phosphorylation observed at 2 s was only slightly, if at all,
influenced by the lipophilic ions. The experiments performed in the
absence of lipophilic ions (Fig. 12) showed, however, that: 1)
in the presence of low NaCl (1 and 3 mM) an initial high
formation of EP was followed by a decrease in EP to a
steady state level, which was reached after more than 1 min (not
shown), 2) the size of the ``overshoot,'' i.e. the
difference between the 2-s value and the steady state level of EP, decreased with increasing [NaCl], and, at 10
mM NaCl, no overshoot was observed, and 3) TPB
Figure 13:
Na-ATPase activity as a function of
[NaCl]. The experiments were performed with bovine brain
Na,K-ATPase as described under ``Materials and Methods.'' Empty circles, activity in the presence of NaCl. Panel
A, effect of 5 (empty triangles) or 10 (filled
triangles) µM TPB
When the measurements were repeated in the presence of 20 mM KCl, the maximal hydrolysis rate was increased by a factor of
Figure 14:
Na,K-ATPase activity. The experiments
were performed with bovine brain Na,K-ATPase as described under
``Materials and Methods.'' Panel A, effect of
K
In the second set of experiments (Fig. 15), both
NaCl and KCl were present, and the sum of [NaCl] and
[KCl] was kept equal to 150 mM. The experiments were
carried out without or with 100 µM ATP. The effect of
TPP
Figure 15:
Effect of K
In another series of experiments, we
investigated whether the effect of TPP
Figure 16:
Effect of sequential incubations with
TPP
The first question to consider is whether the lipophilic ion
effect on the Na,K-ATPase reaction is due to a modification of the
intramembrane electric field. The experiments with sequential additions
of lipophilic ions to the medium (Fig. 16) indicate that the
effects of TPP In the present
study, our attention was focused on the reaction steps where
Na
Despite a significant
TPP
Under conditions where the K-sensitive EP amounted to 86-90% of the EP level, the rate
coefficient of Na Lipophilic ions affected
the rate coefficient for both the Na
To evaluate whether this observation entailed
a TPP
The reaction sequence does not include K-insensitive EP
comprising, at 100 mM NaCl, less than 10% of the
phosphoenzyme. When k The rate coefficient for the rapid decay, k
The conclusion about the effect of lipophilic
cation on the reactivity of the K-sensitive EP toward the
extracellular cations is supported by a similar evaluation of the
spontaneous, Na Thus, based on the analysis of both the
Na
The overshoot observed in the
absence of K
At saturating substrate concentrations, the effect
of K
At
low concentrations of KCl, the activity was strongly inhibited by NaCl
(not shown), but the inhibition was almost completely reversed by low
[ATP] (Fig. 15). As previously suggested by Post et al.(1972) and by Drapeau and Blostein(1980), this
activating effect of ATP might be caused by an increase in the
concentration of E
Our results are in
accord with this hypothesis. Lipophilic ions affect the interaction of
Na But could the effect of TPP Interaction between the
introduced charge in the membrane and charged residues in the protein
near or inside the cation binding sites could result in some
modification of the protein structure. In this case, one would expect
that lipophilic cation and anion would modify different charged
residues. The fact that TPP
Volume 270,
Number 9,
Issue of March 3, 1995 pp. 4244-4254
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
) and tetraphenylboron (TPB
), on
interactions of Na and K with
Na,K-ATPase were studied with membrane-bound enzyme from bovine brain,
pig kidney, and shark rectal gland. and
K interactions with the inward-facing binding
sites, monitored by eosin fluorescence and phosphorylation, were not
influenced by lipophilic ions.-, ADP-, and Na-dependent
dephosphorylation. TPP decreased: 1) the rate of
transition of ADP-insensitive to ADP-sensitive phosphoenzyme, 2) the
K affinity and the rate coefficient for
dephosphorylation of the K-sensitive phosphoenzyme, 3) the
Na affinity and the rate coefficient for
Na-dependent dephosphorylation. Pre-steady state
phosphorylation experiments indicate that the subsequent occlusion of
extracellular cations was prevented by TPP.
TPB had opposite effects. was replaced by N-methyl-D-glucamine or Tris, but
were unaffected by the replacement of Cl by other
anions. and
K access to their extracellular binding sites is
modified by the intramembrane electric field.
) (
)and tetraphenylboron
(TPB) partition into cell membranes (Andersen et
al., 1978; Flewelling and Hubbell, 1986a, 1986b). The ions are
located a few angstroms below the membrane surface (Andersen et
al., 1978), and they modify the electrostatic potential profile
inside the membrane dielectric (Andersen et al., 1978;
Flewelling and Hubbell, 1986a; 1986b). It has been shown by
Bühler et al.(1991),
Läuger (1991a), and Stürmer et al.(1991) that these ions affect the reactivity of
membrane-bound Na,K-ATPase toward Na and
K. The authors concluded from experiments with
fluorescent, potential-sensitive aminostyrylpyridinium dyes that
TPP and TPB affect the release of
Na from and the binding of K to
external sites of Na,K-ATPase. The authors therefore suggested that the
external cation binding sites are formed as deep wells, i.e. K binds to and Na is released
from ``sites that are located inside the membrane
dielectric'' (Läuger, 1991a). Since the
binding of Na at the cytoplasmic side of Na,K-ATPase
caused relatively small changes in the fluorescence response, the
authors concluded that the cytoplasmic sites are located close to the
lipid/water interface (Läuger, 1991a).
denotes
the form of Na,K-ATPase characterized by a high affinity to ATP and
Na, while E
denotes the form with
a high affinity to K but a relatively low affinity to
ATP and Na (Glynn, 1985; Läuger,
1991b). The lower row of the scheme shows reaction of Na and K with the cytoplasmic sites, while the
upper row illustrates reaction of cations with the extracellular sites.
The phosphoenzyme EP(Na
) has m occluded Na ions and reacts with ADP
resynthesizing ATP. EP binds K with a very high affinity, dephosphorylates, and forms an
intermediate with n occluded K-ions, E(K
).(K) and ENa were characterized using
eosin fluorescence as an indicator of conformation (Skou and Esmann,
1983); the steps leading to formation of phosphoenzymes were studied in
experiments where the pre-steady state time course of E
P formation was measured, and the individual
phosphoenzyme forms were characterized in chase experiments with ADP or
K. The entire reaction cycle was characterized through
measurements of the steady state (1) phosphorylation level, (2) distribution of ADP- and K-sensitive phosphointermediates, (3) ATP and pNPP hydrolysis activity at varying concentrations
of Na or K or both. The effects of
lipophilic ions on the overall reaction are compared to their effects
on the partial reactions. Preliminary results have been published
(Klodos & Plesner, 1992).
Enzyme Preparation
Na,K-ATPase from bovine brain
was prepared according to Klodos et al.(1975). Na,K-ATPase
from pig kidney was prepared according to Jørgensen(1974) as
modified by Jensen et al.(1984). Na,K-ATPase from shark rectal
gland was prepared as described by Skou and Esmann(1979).
(mg of protein)
, respectively.
The maximum phosphorylation site concentrations, measured as described
(Klodos et al., 1981), were 0.45-0.55 nmol(mg of
protein) for the brain enzyme, 2.7 nmol(mg of
protein) for the pig kidney enzyme, and 2.5
nmol(mg of protein) for shark rectal gland
Na,K-ATPase.Measurement of Eosin Fluorescence
The experiments
were performed in a SPEX Fluorolog-2 spectrofluorometer at 20 °C,
the excitation wavelength was 530 nm, the slit was 0.1 nm, and the
light path was 1 cm. A Schott RG550 filter was used as emission cut-off
filter. and subsequently to addition of
Na was measured in 10 mM HEPES, 10 mM MES, 10 mM EDTA buffer, pH 7.5, which favors a form of
Na,K-ATPase with high affinity to eosin. The pH was adjusted with N-methyl-D-glucamine. The fluorescence of
unspecifically bound eosin was measured in the presence of 375
µM ADP, which prevents specific binding of eosin (Esmann,
1992).
= 0.05%). The pig kidney
Na,K-ATPase (30 µg of proteinml) in 2 ml
of buffer solution was thermostated and continually stirred. All
additions were from hand-held Hamilton syringes. For further analysis,
data were expressed in percent of the starting level of fluorescence in
each individual experiment.Phosphorylation-Dephosphorylation Experiments
The
experiments were performed at 0 °C and pH 7.4 with 300 µg of
proteinml. TPP and
TPB, when present, were added from freshly prepared
solutions in water at least 10 min prior to the start of the
phosphorylation. The phosphorylation medium also contained 0.1 mM MgCl, 30 mM imidazole buffer, pH 7.4, at 0
°C and varying [NaCl]. The phosphorylation was started by
the addition of 25 µM [
-P]ATP.
After varying periods of time, the reaction was either stopped by
addition of 10% trichloroacetic acid (final concentration 5%) or
dephosphorylation was initiated by the addition of a chase. The chase
contained NaCl, buffer, and MgCl in the same concentrations
as the phosphorylation medium and, in addition, either unlabeled ATP
(final concentration 1 mM), unlabeled ATP and KCl (final
concentrations were 1 mM and usually 20 mM respectively), or unlabeled ATP and ADP (final concentrations 1
and 2.5 mM). The dephosphorylation was stopped by addition of
10% trichloroacetic acid (final concentration 5%) at the times
indicated. The amount of acid-stabile EP was
determined according to Klodos et al.(1981), and the results
are presented in the figures after subtraction of EP levels,
obtained after dephosphorylation for 5 min. The values in the figures
are the mean of at least three experiments ± S.E.Measurement of ATPase Activity
The
[-P]ATP hydrolysis was measured in 30
mM histidine buffer, pH 7.4, at 37 °C in the presence of 3
mM ATP, 3 mM MgCl, and [NaCl]
and [KCl] given in the figures. P
was determined according to Lindberg and Ernster(1956).Measurement of pNPP Hydrolysis
The pNPPase
activity was measured according to Ottolenghi(1975), but without bovine
serum albumin in the incubation medium. The buffer was 30 mM imidazole buffer, pH 7.5, at 37 °C. The reaction medium
contained 20 mM MgCl and 10 mM pNPP. The
pNPPase activity was measured as function of [KCl] at a
constant salt concentration of the medium, 150 mM, replacing
KCl with choline chloride, NMGCl, or NaCl. Identical results were
obtained with choline chloride and NMGCl. The experiments in the
presence of both KCl and NaCl were performed with or without 100
µM ATP. The amount of the released p-nitrophenol
was determined from the optical density at 410 nm.Data Processing
The data were analyzed using the
computer program ``Plot 5.31'' written by Bliss Forbush III,
Dept. of Cellular and Molecular Physiology, School of Medicine, Yale
University, applying a linear or nonlinear least squares analysis.Reagents
ATP and ADP were purchased as sodium
salts from Boehringer Mannheim, Germany. ATP used in the
phosphorylation experiments was converted to its Tris salt by
chromatography on a Dowex 1 column (from Sigma), and
[-P]ATP was purified on DEAE-Sephadex G-25
(Nørby and Jensen, 1971). ADP was purified by chromatography on
a Dowex 50W H column, and the eluate was adjusted to
pH 7.1 at 20 °C with 2-amino-2-methyl-1,3-propanediol. p-Nitrophenyl phosphate (pNPP) purchased as sodium salt from
Merck, Darmstadt, Germany, was purified by chromatography on Dowex 50W
H. The eluate was adjusted with Tris to pH 7.4 at 37
°C. N-Methyl-D-glucamine (NMG) was purchased from
Sigma. Eosin was obtained from Gurt, Chadwell Heath, Essex, UK. TPPCl
and NaTPB were gifts from Dr. H.-J. Apell,
Universität Konstanz, or purchased from Sigma.
TPPCl and NaTPB were dissolved in dimethyl sulfoxide or in water. The
water solutions of TPPCl and NaTPB were prepared on the day of the
experiment. All other reagents were reagent grade.
and K to
the dephosphoenzyme, phosphorylation, characterization of
phosphointermediates, Na,K-ATPase and pNPPase activities, and, finally,
a series of experiments aiming at clarification whether the effects of
lipophilic ions are due to their charge. It should be noted that: 1)
because of its higher partitioning into the lipid phase (Flewelling and
Hubbell, 1986a, 1986b) TPB was used in lower
concentrations than that of TPP, and 2) whenever
K was present only the effect of TPP was tested, as TPB is a strong chelator of
K (Flaschka and Barnard, 1960).Interconversions between Dephosphoenzyme Forms:
E
Conformational transitions of the Na,K-ATPase
following cation binding to nonphosphorylated enzyme forms were
monitored by eosin fluorescence according to Skou and Esmann(1983), who
showed that the fluorescence of eosin specifically bound to the enzyme
is higher than the fluorescence of eosin in the solution or of
nonspecifically bound eosin. Specifically bound eosin was released from
its high affinity binding site by the addition of ADP (Esmann, 1992).
375 µM ADP used in the reported experiments was more than
sufficient to prevent specific eosin binding. The fluorescence level
observed in the presence of ADP was equal to that measured in the
presence of saturating KNa
E(K
)
Transition, where E(K), which did not bind eosin
specifically, was the only form present. Neither TPP nor TPB had any effect on the affinity of
specific eosin binding measured as equilibrium eosin binding (not
shown). Both ions affected eosin fluorescence: TPP caused an increase in the fluorescence of nonspecifically bound
eosin, i.e. it increased the background fluorescence, while
TPB increased the quantum yield of the fluorescence
only of specifically bound eosin (Fedosova and Jensen, 1994). did not result in an increase in fluorescence, indicating that
already in the absence of Na the specific eosin
binding was maximal. The difference between the level of fluorescence
in the presence of buffer, Na, or both, and the level
in the presence of ADP was equal to the maximal fluorescence increase
caused by the eosin binding to the enzyme
(
F
). F corresponded therefore to the complete transition of the enzyme
from eosin-bound E to the eosin-free E form.,
the addition of K induced a decay of specific
fluorescence which could be fitted adequately with a monoexponential
function (Fig. 1, inset). The rate coefficient of the
decay (k
) and the equilibrium fluorescence
decrease (F/F
) were derived.
F/F versus
[K] was found to be a hyperbola:
F/F =
(F/F)/(1 + K
/[K]),
where K = 5.2 ± 0.9
µM (not shown). Similar values were previously found in
different types of experiments (cf. Esmann, 1992). The rate
coefficient of fluorescence decrease, k, in the
same range of potassium concentrations, showed a linear dependence on
[K] (Fig. 1A) indicating a
low affinity K binding.
concentrations on the rate coefficient of E 
E conversion. Inset in panel A shows a typical recording of eosin
fluorescence in an experiment performed with pig kidney Na,K-ATPase in
the presence of buffer. Ligand additions are shown and fluorescence is
expressed as percentage of the initial level. Thick line shows
monoexponential fit of the fluorescence decrease. The calculated rate
coefficients for the KCl-dependent fluorescence decrease are shown in panels A and B: panel A, without NaCl; panel B, with 4 mM NaCl in the absence (circles) or in the presence of 300 µM TPP (squares). The values are the mean
of three experiments ± S.E. Note the 100-fold difference in
concentrations between panel A and panel
B.
binding to
the E form is followed by a conformational
transition, poised heavily in favor of the E form (cf. reviews by Glynn(1985) and Glynn and Karlish (1990)). The
rate coefficient of disappearance of the E form
upon K addition is equal to k = k
+ k
/(1 + K
/[K]) and becomes k = k +
k[K]/K for [K] 
K.
Glynn et al. (1987) showed that the time course of deocclusion
and the conformational change are closely correlated. Thus, k in the model is the rate constant for
K deocclusion from the intracellular sites., k,
displayed a linear dependence on [K] (Fig. 1A). The rate constant k for the E E transition was estimated to be 0.03 ± 0.012 s by extrapolation of the straight line to 0
[K]. It is in agreement with the data on
deocclusion and conformational transition rate constants measured by
different techniques (Glynn, 1985; Glynn et al., 1987). = K(k/k),
and as k is about 300 s (Steinberg and Karlish, 1989) and k = 0.03 s, an approximate value of the
dissociation constant for K, K,
was estimated to be in the millimolar range. was present, i.e. the starting point
was ENa, K addition
was also followed by a monoexponential decay. The rate constant k for the K deocclusion
was unchanged, but the slope of k versus
[K] was decreased by a factor of 80 (Fig. 1B) and the K
value obtained from equilibrium measurements was 421
µM (Fig. 2). The simplest reaction scheme
compatible with these results is
in the presence of 4 mM NaCl.
Eosin fluorescence was measured in 4 mM NaCl, 10 mM HEPES, 10 mM MES, 10 mM EDTA, pH 7.5, and
different concentrations of K in the absence (circles) or in the presence of 300 µM TPP (squares). Eosin fluorescence upon
addition of K is expressed as percentage of the
starting level of fluorescence. The values are the mean of three
experiments ± S.E. The curve is a fit of the experimental data
to the equation: F/F =
(F/F)/(1 + K
/[K]),
as described under ``Materials and Methods.'' The K = 421 ±
0.9 µM.
and K compete for the binding to the E form, but
neither the K occlusion, defined as EK to EK transition, nor
deocclusion are affected by Na. TPP had no effect on the K binding in the presence
of Na (Fig. 1B and Fig. 2). binding in the absence of
K were performed at low buffer concentration (1 mM HEPES + 1 mM MES) where no specific eosin binding
was observed. It has been previously shown that in low ionic strength
medium the enzyme has the characteristics of an E conformation (Skou and Esmann, 1980; Glynn and Richards, 1982;
Jensen and Ottolenghi, 1983b; Klodos and Ottolenghi, 1985).
Na addition to the medium was followed by a
fluorescence increase which reflected the formation of the sodium-bound
form of the enzyme. Neither TPP nor TPB affected the Na dependence of the specific
fluorescence increase (not shown). Thus, TPP affected
neither Na nor K binding nor
K deocclusion.Formation of Phosphoenzyme
The amount of EP was measured after a 2-s phosphorylation period at NaCl
concentrations varying from 0 to 100 mM. We reported
previously a marginal effect of the lipophilic ions on the formation of
phosphoenzyme under these conditions ( Fig. 2in Klodos and
Plesner(1992)). decreased the apparent affinity to Na, and the
Na dependence curve became S-shaped. The
half-saturating [NaCl] in the presence of TrisCl was 2.7
± 0.12 mM (n = 6), and only 0.33
± 0.02 mM (n = 6) with NMGCl in the
medium. With 100 mM NMGCl, in the absence of added NaCl,
4-12% of the enzyme was phosphorylated (not shown), but it is not
clear whether this phosphorylation was caused by NMG itself or by traces of Na.
and
TPB on the formation of phosphoenzyme. Bovine brain
Na,K-ATPase was incubated for 10 min at 20 °C in 30 mM imidazole buffer (pH 7.4 at 0 °C), 0.1 mM MgCl, and varying [NaCl] and
[NMGCl] (left panel) or [NaCl] and
[TrisCl] (right panel), in the absence of lipophilic
ions (circles) and in the presence of 300 µM TPP (squares) or 33 µM TPB (triangles). The sum of
[NaCl] and [NMGCl] or [TrisCl] was 100
mM. The samples were subsequently cooled to 0 °C. 25
µM [-P]ATP (final) was added,
while concentrations of other components remained unchanged. The enzyme
was phosphorylated at 0 °C. After 2 s, the phosphorylation was
stopped and the amount of phosphoenzyme was measured as described under
``Materials and Methods.'' The figure shows the amount of
phosphoenzyme formed at 0-10 mM NaCl.
Na-dependent
Dephosphorylation
Na-dependent
dephosphorylation of the phosphoenzyme was examined in the absence of
ADP or KCl. The experiments were performed at 10-300 mM NaCl, and the chase contained 1 mM unlabeled ATP. The
Na dependence of the dephosphorylation rate
coefficients, obtained by a monoexponential fit of the
dephosphorylation data, is shown in Fig. 4.
(squares) or 33 µM TPB (triangles). At zero time, a chase containing unlabeled
ATP (1 mM final) was added. The dephosphorylation was stopped
and the first order rate constants were estimated as described under
``Materials and Methods.''
-insensitive EP (less than 12% of the total EP, Fig. 9), neither of the two processes could
significantly affect the dephosphorylation of this phosphoform. Thus,
NaCl stimulation of the Na-dependent dephosphorylation
was due to Na acting as a replacement, albeit poor,
for K in activating the dephosphorylation of
K-sensitive EP. As shown in Fig. 4, the rate
coefficient of Na-dependent dephosphorylation was
increased by 33 µM TPB at
[NaCl] lower than 150 mM, whereas it was decreased
by 300 µM TPP at all NaCl concentrations.
effect on
K-stimulated dephosphorylation of bovine brain
Na,K-ATPase. The experiments were performed as in Fig. 8with
bovine brain Na,K-ATPase at varying [NaCl] in the absence (circles) or in the presence of 300 µM TPP (squares). [KCl] in the
chase was 20 mM (final). A biexponential function EP
= EP
exp(-k
t)
+ EP
exp(-k
t)
was fitted to the data like those in Fig. 8as described under
``Materials and Methods.'' The relative amount of
K-insensitive phosphoenzyme, EP, the rate
coefficient of the decay of EP
, k, and the rate coefficient of the decay of EP
, k, are shown as
function of [NaCl] in panels A, B, and C,
respectively.
. The
phosphorylation was performed as in Fig. 4with bovine brain (upper row) or pig kidney Na,K-ATPase (lower row) in
the presence of 100 mM NaCl without (left column) or
with 300 µM TPP (right column).
At zero time, a K chase containing unlabeled ATP (1
mM final) and [KCl] (final), 1 mM (triangles), 20 mM (squares), or 50
mM (circles), was added. Dephosphorylation was
stopped at the times shown.
ADP-dependent Dephosphorylation
Neither
TPP nor TPB in concentrations up to
600 µM and 100 µM, respectively, influenced
the steady state EP level at [NaCl] 
10 mM (not shown). In a series of experiments, EP was formed in
the presence of 10 to 500 mM NaCl with or without lipophilic
ions for 60 s and subsequently chased with 2.5 mM ADP.
(squares) or 33
µM TPB (triangles). At zero
time, a chase solution containing unlabeled ATP (1 mM final)
and ADP (2.5 mM final) was added. Dephosphorylation was
stopped at the times shown. Panel A, dephosphorylation time
course at 100 mM NaCl. A biexponential function EP = EP
exp(-kt)
+ EP
exp(-kt)
was fitted to the data like those in panel A as described
under ``Materials and Methods.'' EP, the relative amount of ADP-sensitive
phosphoenzyme, and k, the rate coefficient of
decay in the slow phase, are shown as functions of [NaCl] in panels B and C,
respectively.
is a rapidly and EP
is a slowly decaying phosphoenzyme
form(s). EP and EP are not synonymous with the classical ADP-sensitive,
K-insensitive EP or the ADP-insensitive,
K-sensitive EP, but signify operational quantities
of phosphoforms characterized by their, rapid or slow, decay. k
is the rate coefficient of ADP-dependent decay, k
and k
are the rate
constants of forward and backward transitions between the ADP-sensitive
and the ADP-insensitive phosphointermediates, and k
is the rate coefficient of dephosphorylation of the
ADP-insensitive EP. The forward transition is accompanied by
dissociation of at least 1 Na (Yoda and Yoda, 1987;
Glynn, 1988; Jørgensen, 1991, 1994; Goldshleger et al.,
1994). Both the backward transition, accompanied by the binding of NaCl
(Post et al., 1975; Nørby et al., 1983;
Nørby and Klodos, 1988), and the dephosphorylation of the
ADP-insensitive phosphointermediate, reflected by the rate coefficient
of the Na-dependent dephosphorylation (Fig. 4)
(Klodos et al.(1981); cf. Nørby and Klodos
(1988)), are dependent on [NaCl]. The rate coefficient of the
decay in the slow phase is equal to the sum of k and k. caused a large decrease in both the steady
state amount of ADP-sensitive EP and in the rate of the slow
phase at all NaCl concentrations (Fig. 5). The opposite was seen
with TPB, which caused a somewhat smaller, but
significant, increase in both the steady state amount of EP and a significant increase in the slope
of the slow phase (Fig. 5). In other words, the effect of
TPP was similar to that of a decrease in the
concentration of NaCl and the effect of TPB to an
increase in [NaCl].
; open triangles, +33 µM TPB-; or filled triangles, 100 µM TPB-.
showed a more pronounced effect than 33 µM (Fig. 6). With TPP in the medium, the
half-maximal effect was obtained at 
30 µM (Fig. 7). In these experiments, the relative amount of the
ADP-sensitive EP was estimated as the amount of phosphoenzyme
removed by a 2-s ADP chase.
]. The experiment was
performed with bovine brain Na,K-ATPase phosphorylated for 60 s in the
presence of 100 mM NaCl and varying
[TPP]. An ADP chase containing unlabeled ATP
(1 mM final) and ADP (2.5 mM final) was added after
60 s of phosphorylation, and the dephosphorylation was stopped after 2
s as described under ``Materials and Methods.'' The fraction
of ADP-sensitive phosphoenzyme is defined here as the proportion of
phosphoenzyme removed by the ADP chase of 2-s duration,
EP
= EP - EP, and shown as percentage of EP.
K-dependent Dephosphorylation
TPP effect on the K-dependent dephosphorylation was
tested in experiments performed with bovine brain, pig kidney, and
shark rectal gland Na,K-ATPases. showed much lower
apparent affinity toward K than the enzyme
phosphorylated under the same conditions but in the absence of
TPP (Fig. 8). In the absence of
TPP, identical dephosphorylation was observed with a
K chase containing 20 or 50 mM KCl, and 1
mM KCl was almost as efficient as 20 or 50 mM KCl (Fig. 8, left column). This is clearly not the case in
the presence of 300 µM TPP (Fig. 8, right column), where 1 mM KCl
was far from saturating. 20 mM KCl was enough to elicit
maximum dephosphorylation in the presence of lipophilic cation, and,
therefore, in the experiments described below, the K chase contained 20 mM KCl.
and EP are not synonymous with the classical EP or EP, but signify operational quantities of
different phosphoforms characterized by their decay. The rapid phase of
dephosphorylation corresponds to the K-sensitive phosphoenzyme, EP, decaying with the apparent rate constant k
(the rate coefficient in the presence of
K), while the slow phase corresponds to the
K-insensitive phosphoenzyme, EP. The rate
constant of the slow decay is equal to the sum of forward transition
rate coefficients, k
and k
. in the
medium. The relative amount of K-insensitive EP increased with
an increase in [NaCl] as expected from the previously
published data (Hara and Nakao, 1981; Nørby et al.,
1983; Klodos et al., 1994). The rate coefficient for the rapid
phase (Fig. 9C) showed an inhibition of the rapid decay
by high [NaCl]. This inhibition was not caused by a
competition between NaCl and K, since even at 600
mM NaCl an increase in [KCl] from 20 to 50 mM did not affect the rapid phase of the dephosphorylation (not
shown). When the phosphorylation was performed in the presence of 300
µM TPP, the rate coefficient of the rapid
phase (Fig. 9C) decreased at all [NaCl]
tested, while k, the rate coefficient of decay
of the K-insensitive EP neither depended on [NaCl]
nor did it change in the presence of TPP. Similar
effects of TPP on the K-dependent
dephosphorylation were observed with pig kidney (Fig. 8) and
shark rectal gland Na,K-ATPase (not shown).Is Na
It has been shown previously that both the
steady state ratio of the phosphoenzyme forms and their
dephosphorylation kinetics are modified by high concentrations of salt
(Post and Suzuki, 1991; Klodos et al., 1994). We tested,
therefore, whether the effects of lipophilic ions were modified by
various salts. The experiments were performed either at 100 mM [Cl Essential for the Effects of
Lipophilic Ions?], substituting most of
Na by NMG, or Tris,
or at 100 mM [Na], replacing most
of Cl by nitrate, acetate, or sulfate (not shown).
The results of the ADP and the K chase experiments are
shown in Fig. 10and 11, respectively, where the experiments at
10 mM NaCl alone are shown for comparison .
,
or 90 mM CHCOONa in the absence (empty
bars) or in the presence of 300 µM TPP (light shadowed bars) or 33 µM TPB (heavily shadowed bars). The data
were evaluated as in Fig. 5. Panel A, EP, the relative amount of ADP-sensitive
phosphoenzyme; panel B, k, the rate
coefficient of decay in the slow phase.
by other cations or substitution of Cl by other
anions caused small, but significant, changes in the relative amounts
of the ADP-sensitive and of the K-insensitive phosphoenzymes. The
changes were as expected from the literature, indicating that the
lyotropic effects of salts on the distribution of phosphoenzymes appear
already at salt concentrations as low as 100 mM.
dependent dephosphorylation. The experiments
were performed as in Fig. 9and Fig. 10. The data were
evaluated as in Fig. 9. Panel A, EP, the relative amount of K-insensitive
phosphoenzyme; panel B, k, the rate
coefficient of decay of the K-sensitive phosphoenzyme. Empty bars in the absence and shadowed bars in the presence of 300
µM TPP.
-induced decreases in
the rate coefficient of the rapid decay of K-sensitive EP (Fig. 11). Moreover, the replacement of Cl for
other anions did not modify the ADP-dependent dephosphorylation in the
presence of lipophilic ions (Fig. 10). The only modification of
the effect of the lipophilic ions on the phosphoenzyme distribution and
the ADP-dephosphorylation kinetics was seen when Na was replaced by other cations. Although the relative amount of
ADP-sensitive EP in the presence of 10 mM NaCl and 90
mM NMGCl was significantly higher than with 10 mM NaCl alone, the effect of lipophilic ions on the phosphoenzyme
distribution remained the same as with 10 mM NaCl without
NMGCl. The lipophilic ions effect on the distribution of phosphoenzymes
disappeared completely in the presence of Tris (Fig. 10A). The rate coefficient of the decay of
the ADP-insensitive EP was decreased by both Tris and NMG in the absence of lipophilic ions, but
returned to a value characteristic for 10 mM NaCl in the
presence of TPB (Fig. 10B).
TPP, however, did not decrease the coefficient to the
same value as with 10 mM NaCl + TPP but
to a value measured with lipophilic cation at 100 mM NaCl. At
present, we have no explanation of this difference in NMG or Tris modification of the effect of
TPP on the distribution of phosphoenzymes on one hand,
and on the rate coefficient for the slow phase of the ADP-dependent
dephosphorylation on the other.Phosphoenzyme as Function of Time at Low
[NaCl]
We observed that although the two lipophilic
ions did not affect the steady state EP level at
[NaCl] higher than 10 mM, they both showed a
pronounced effect on the steady state level at lower [NaCl].
The last observation seemed to be in conflict with the results of 2-s
phosphorylation (Fig. 3). To elucidate this apparent
controversy, we measured the time course of EP formation at 1,
3, and 10 mM NaCl and 33 µM TPB or 300 µM TPP. The results are
shown in Fig. 12.
(squares) or 33 µM TPB (triangles). 100% corresponds to
the steady state value obtained at 100 mM NaCl.
increased the overshoot by decreasing the steady state levels at
1 and 3 mM NaCl. In contrast, 4) in the presence of 300
µM TPP, no overshoot was observed at any
[NaCl].ATPase Activity
The rate of ATP hydrolysis in the
absence of K at 3 mM ATP, 3 mM MgCl increased with [NaCl] up to about 400
mM, whereas inhibition was caused by higher NaCl
concentrations (Fig. 13). A concentration-dependent shift of the
curve to the right was observed with TPP (2 and 5
µM), i.e. higher concentrations of Na were needed for activation as well as for inhibition.
TPB (5 and 10 µM) had the opposite
effect. (The concentrations of lipophilic ions in these experiments
were decreased because of the very low protein concentrations used.)
on the activity. Panel B, effect of 2 (empty squares) or 5 (filled
squares) µM TPP.
25 and was obtained at [NaCl] 150 mM (Fig. 14). The effect of lowering [KCl] was to
decrease the maximal hydrolysis rate and to shift the curve to the
left, i.e. the maximal hydrolysis rate was obtained at a lower
[NaCl] (Fig. 14A). The effect of 20
µM TPP was equivalent to decreasing the
concentration of K (Fig. 14B).
on the activity. The activity was measured as a
function of [NaCl] in the absence of KCl (empty
circles) or with 2.5 mM KCl (filled circles), 5
mM KCl (empty squares), or 20 mM KCl (filled squares). Panel B, effect of TPP on the activity in the presence of 5 mM KCl. The
activity in the absence of TPP, empty
squares, and with 20 µM TPP, empty circles.
pNPPase Activity
As shown in Fig. 8and Fig. 9, TPP affected very strongly the
properties of the K-sensitive EP. To further elucidate the
interaction of the enzyme with K, we measured the
activity of K-stimulated pNPPase, which is assumed to
reflect potassium-sensitive steps in the Na,K-ATPase reaction mechanism (cf. review by Glynn(1985)). The activity was measured as a
function of [KCl] in two series of experiments. The first set
was performed with varying [KCl], and the ionic strength was
kept constant with choline or NMG chloride. Under these conditions,
TPP had very little, if any, effect on the activity
(not shown). was clearly seen both in the presence and in the
absence of ATP (Fig. 15). With ATP in the medium, TPP caused both a decrease in the apparent affinity for KCl and a
change in the shape of the K activation curve.
and 100
µM ATP on pNPPase activity in the presence of both
Na and K. The experiments were
performed at 37 °C with bovine brain Na,K-ATPase as described under
``Materials and Methods.'' The medium contained 20 mM MgCl, 10 mM pNPP, 30 mM imidazole
buffer, pH 7.5, at 20 °C, KCl and NaCl. The sum of [NaCl]
and [KCl] was equal to 150 mM. The experiments were
performed in the absence (left column) or in the presence (right column) of 100 µM ATP. Squares,
activity in the presence of 300 µM TPP. Lower row, expanded abscissa.
Is the Effect of Lipophilic Ions Caused by Their
Charge?
Our original assumption that the effects of lipophilic
ions were caused by their charge required an experimental proof. To
rule out the possibility that the observed effects were due to an
irreversible protein modification by lipophilic ions, we examined
whether the effect of TPP would disappear after
removal of the membrane-bound TPP. In these
experiments, the enzyme was incubated with 100 or 300 µM TPP and subsequently repeatedly washed by
centrifugation and resuspension in TPP-free medium. As
a control, TPP-bound enzyme was incubated and washed in a medium
containing TPP, and TPP-free enzyme was exposed to the
same treatment but in the absence of TPP. After the
subsequent phosphorylation, the phosphoenzymes were probed with a 2-s
2.5 mM ADP chase or a 2-s 1 mM K chase, the concentration of KCl at which the TPP modification of the phosphoenzyme behavior was most apparent (Fig. 8). The enzyme, pre-exposed to TPP and
subsequently washed in TPP-free medium, and the
TPP-free enzyme formed identical phosphointermediates
with respect to their K and ADP sensitivity (not
shown), thus excluding an irreversible modification of the protein by
TPP. was reversed by
a subsequent addition of TPB and vice versa (Fig. 16). In these experiments, the enzyme was incubated
in the phosphorylation medium containing 100 mM NaCl and 50 or
300 µM TPP. After 15 min, 3.3 or 33
µM TPB was added, and the incubation was
allowed to proceed for another 15 min (Fig. 16, left
panel). In a parallel series of experiments, TPB was present during the first 15 min of incubation and
TPP was included during the next 15 min (Fig. 16, right panel). The enzymes were then
phosphorylated and the phosphoenzymes were probed with a 2-s ADP chase.
It is obvious from Fig. 16that the effect of one lipophilic ion
was reversed by a lipophilic ion of the opposite charge. The phenomenon
was concentration-dependent but was independent of the sequence of
additions, i.e. of the charge of lipophilic ion that was first
``seen'' by the enzyme. Thus, our conclusion from these
experiments is that the effect of lipophilic ions on the properties of
the enzyme is related to their electric charge.
and TPB on ADP sensitivity of
phosphoenzyme. Left panel, bovine brain Na,K-ATPase was
incubated at 20 °C in a medium containing 100 mM NaCl, 0.1
mM MgCl, and 30 mM imidazole buffer in
the presence of 50 or 300 µM TPP for 15 min. Subsequently,
0, 3.3, or 33 µM TPB was added for an
additional 15 min. Right panel, the enzyme was incubated first
for 15 min with 3.3 or 33 µM TPB, then
0, 50, or 300 µM TPP was added for the
following 15 min. After the incubation, the samples were cooled to 0
°C and phosphorylated for 60 s at 100 mM NaCl, 0.1 mM MgCl, and 25 µM [-P]ATP. The ADP sensitivity was
probed as in Fig. 7, and the fraction of phosphoenzyme remaining
after a 2-s chase is shown. The same experiment performed in the
absence of lipophilic ions is shown for comparison (labeled ``100
mM NaCl'').
and TPB are caused by
their charge. We ascribe therefore the effects of lipophilic ions to
modifications of the electric field in the membrane. and K are known to bind to or
leave the enzyme. To discriminate between Na and
K interactions with intra- and with extracellular sites, we studied partial reactions involving
dephosphoforms, known to have their inward-facing sites open,
and phosphoenzymes, where the outward-facing sites are
accessible (cf. reviews by Glynn and Karlish(1990), Robinson
and Pratap(1993), and Vasilets and Schwarz(1993)). An attempt was also
made to correlate the induced changes in the partial reactions with the
effects on the overall reaction cycle.K
Na and Na Interaction with
Intracellular Sites and K binding to Na,K-ATPase and transitions between dephosphoenzymes, E(K) and ENa, were investigated using eosin as
an indicator of conformations (Skou and Esmann, 1983). As documented
under ``Results,'' neither affinity toward Na nor K were affected by the lipophilic ions
implying that the binding of Na or K to the E form of the enzyme and the
occlusion of intracellular K are not influenced by the
intramembrane electric field. Moreover, the formation of phosphoenzyme,
which is a function of Na binding to the intracellular
sites, was not affected by the lipophilic ions. Thus, the interaction
of Na,K-ATPase with the intracellular Na and
K appears to be independent of the intramembrane
electric field. This conclusion is in agreement with the results
obtained with Na,K-ATPase reconstituted into lipid vesicles (Rephaeli et al., 1986a).Na
Phosphorylation of Na,K-ATPase from ATP
leads to a closure of inward-facing cation binding sites and to an
opening of outward-facing cation binding and release sites (cf. reviews by Glynn and Karlish(1990) and Robinson and
Pratap(1993)). Na and K Interaction
with Extracellular Sites release to the external medium
accompanies the transition of ADP-sensitive to potassium-sensitive
phosphointermediates (Yoda and Yoda, 1987; Glynn, 1988;
Jørgensen, 1991, 1994; Goldshleger et al., 1994). The
effect of lipophilic ions on the composition of the phosphoenzyme pool,
the (spontaneous) Na-dependent dephosphorylation, and
the reactivity of phosphoenzymes toward K and ADP in
the K and ADP chase experiments will thus reveal
modifications in the properties of extracellular cation binding sites.Steady State Fraction of ADP- or K-sensitive
Phosphoenzymes
Lipophilic ions affected the steady state
composition of the EP pool, but they affected neither the
formation of EP nor, at NaCl concentrations higher than 3
mM, the steady state level of EP. The ratio between
the ADP-sensitive and the ADP-insensitive EP is determined by
the dissociation of Na or salt from the ADP-sensitive EP and the rebinding of Na or salt to the
ADP-insensitive phosphoform (see Klodos et al., 1994). The
steady state ratio was decreased by TPP (Fig. 5, Fig. 6, and Fig. 7) and increased
by TPB ( Fig. 5and Fig. 6) and,
although the EP ratio was affected by various salts, the
modification of the ratio by lipophilic ions required the presence of
Na (Fig. 10).-induced decrease in the fraction of ADP-sensitive EP, the proportion of K-insensitive EP seemed to be
only slightly decreased by TPP ( Fig. 8and Fig. 9). However, the relative amounts of the phosphoenzymes
disappearing in the two phases of dephosphorylation correspond to the
initial, steady state amounts of these phosphoforms only when the rate
of dephosphorylation for one of the intermediates is much greater than
the rate of transition between phosphoforms (k, k, and k
in )
(Klodos et al., 1981). Thus, it is to be expected that the
experimentally determined amount of K-insensitive EP is larger
in the presence of TPP, since this compound decreases
the rate coefficient for the rapid decay of K-sensitive phosphoform,
but leaves the transition rate coefficients unaffected (see Fig. 9, B and C) This expectation is supported
by the fact that even in the absence of TPP the
fraction of K-insensitive EP seemed to increase when the rate
coefficient of the rapid phase was decreased by lowering
[KCl] in the chase (to 1 mM instead of 20-50
mM, Fig. 8).Kinetics of Na- or ADP-dependent
Dephosphorylation
The effect of lipophilic ions on the steady
state ratio of the ADP-sensitive to the ADP-insensitive EP and
the specific requirement for Na for this effect leads
to a question whether the effect was caused by a modification of the
release of Na from the extracellular site or by
changed rebinding of Na to extracellular sites on the
ADP-insensitive EP. The present data allow us to analyze only
the modification of Na rebinding to the
ADP-insensitive EP, by examining the effect of lipophilic ions
on the kinetics of Na- and ADP-dependent
dephosphorylation.-dephosphorylation reflects only the
forward dephosphorylation of this EP form, k in (see also Klodos et al.(1981)). In the
ADP chase experiments the rate coefficient of the slow phase reflects
both forward dephosphorylation and backward transition of the EP, k and k in .-dependent
dephosphorylation and the slow decay of ADP-insensitive EP.
TPP caused a decrease in both the rate coefficient for
Na-dependent dephosphorylation (Fig. 4) and of
the slow decay (Fig. 5), while TPB induced an
increase in both coefficients. TPB induced a shift of
the Na dependence curve for the
Na-dependent dephosphorylation to lower
[NaCl] (Fig. 4), but did not change the maximal value
of the rate coefficient. At 100 mM NaCl, TPB increased the rate coefficient of the slow decay of the
ADP-insensitive EP (Fig. 5), whereas the rate
coefficient of the Na-dependent dephosphorylation
remained virtually unchanged (Fig. 4). This could indicate that
the TPB-dependent acceleration of the slow decay is
caused by an increase in affinity for Na for the
backward transition of EP to EP. It is, however, not clear whether the
change in the backward transition is sufficient to cause the observed
shift in the ratio of the ADP-sensitive to the ADP-insensitive EP.Kinetics of K-dependent Dephosphorylation
At
saturating [KCl], TPP elicited a decrease in
the rate coefficient of the rapid decay, i.e. the
dephosphorylation of the K-sensitive EP ( Fig. 8and Fig. 9), and the apparent affinity for K (Fig. 8).-induced increase in the dissociation constant, K, of K-bound phosphoenzyme, EPK, the results of experiments in Fig. 8were analyzed according to the scheme
is much lower than both k and k
, the rate
coefficient for the rapid decay k = k/(1 + K/[K]), K being the dissociation constant for
K, k/k. It
is obvious from the equation that the ratio of k at constant nonsaturating [K], i.e. 1 mM KCl, to k, both
determined in the absence or in the presence of lipophilic cation,
should be the same unless K is affected by
TPP., was a saturable function of [KCl]
both in the absence and in the presence of TPP, i.e.k values at 20 mM KCl
were identical with those at 50 mM KCl (Table 1). We
assumed therefore that they represent k. k obtained at 100 mM NaCl in the absence
of lipophilic cation were virtually the same, 1.9 and 2.4
s, for the bovine brain and the pig kidney
Na,K-ATPase, respectively. The corresponding values in the presence of
300 µM TPP were 0.9 s for both enzymes. In the presence of 1 mM KCl in the
chase, the k/k ratio was
about 0.8 for both brain and pig kidney enzyme and decreased to 0.5 in
the presence of TPP (Table 1), implying about a
4-fold increase in K in the presence of
lipophilic cation.
-dependent, dephosphorylation.
TPP decreased significantly the rate coefficient for
Na-dependent dephosphorylation and, less conclusively,
it seemed to decrease the apparent affinity for extracellular
Na (Fig. 4). TPB on the other
hand moved the curve to lower [NaCl] (Fig. 4), thus
indicating an increased apparent affinity for Na binding to the K-sensitive EP's extracellular
cation binding site.-dependent dephosphorylation and the ADP or
K chase experiments, the intramembrane charge alters:
1) Na binding to the ADP-insensitive EP
affecting the backward transition to the ADP-sensitive EP, 2)
K-sensitive EP's affinity toward K or
Na, and 3) the properties of the K-sensitive EP-cation complex, reflected in the dephosphorylation rate
constant. This suggests that the intramembrane electric field affects
Na and K passage to and from the
outward-facing cation binding pocket and alters kinetic properties of EP-cation complexes. A change in the kinetic properties of the
K-sensitive EP-cation complex could reflect a
difference in the way the cation was bound and might also result in
changed properties in the subsequently formed E-cation complex. This appears to be the case,
since TPP seems to prevent occlusion of extracellular
cations.Occlusion of Extracellular Na
Initial formation of phosphoenzyme
in the presence of 1-3 mM NaCl was followed by a
decrease in the amount of phosphointermediate, so-called overshoot (Fig. 12). A generally accepted explanation of such an overshoot
is a decrease of the phosphoenzyme formation due to an accumulation of E and
K(cation
) form. The overshoot was
previously observed with K (Mårdh, 1975; Klodos
and Nørby, 1979), but also in the presence of Na alone, when low ionic strength and low [ATP] did not
support the transition of E(Na) to
the E form (Klodos and Ottolenghi, 1985). In the
latter case, Na acted as K congener
stimulating the dephosphorylation of the K-sensitive EP. The
transition of E(cation) into the
phosphorylatable E form requires either high
[ATP] or, at low [ATP], high [NaCl], or
both, and, in our experiments, with 25 µM ATP in the
medium, 10 mM NaCl was sufficient to prevent the accumulation
of the occluded form (Fig. 12). was increased by TPB,
while in the presence of TPP there was no overshoot.
Since the phosphorylation was not affected by lipophilic cation, the
lack of overshoot in the presence of TPP could be due
to either an increased deocclusion of the E-cation
complex or an inhibition of formation of the E(Na) or both. As lipophilic cation
did not affect transitions between the dephosphoforms, the most likely
explanation is that TPP prevents formation of E(Na). The opposite effect of
TPB on the overshoot indicates that lipophilic anion
increased the occlusion of Na.ATPase Activity
Effect of lipophilic ions on the
interaction of Na and K with the
extracellular sites was also seen in the hydrolytic activity of Na- and
Na,K-ATPase. The bell-shaped Na dependence curve of
Na-ATPase activity might be interpreted as an expression of a shift in
the rate-limiting step from the dephosphorylation of EP, accelerated by [NaCl] 
400
mM, to the EP &rlhar2; EP transition, inhibited by high [NaCl]
acting at a low affinity extracellular binding site (Glynn and Karlish,
1976; Kaplan and Hollis, 1980; Kaplan, 1982). It is in agreement with
this interpretation that 1) with TPP, which slowed
down the dephosphorylation of EP and increased the
rate of transition, the Na dependence curve was moved
to higher [NaCl], i.e. the maximal hydrolytic
activity was obtained at significantly higher [NaCl] and a
much smaller inhibition was observed at very high [NaCl], and
2) TPB, accelerating both the dephosphorylation of EP and its backward transition, decreased the
concentration of NaCl required for the maximal activation of hydrolysis
and increased the degree of inhibition at high [NaCl] (Fig. 13). is to accelerate dephosphorylation and ATP
hydrolysis through binding to the extracellular K site, maximal activity being obtained at 20 mM KCl and
150 mM NaCl. Decreasing under these conditions,
[K] below 20 mM, decreased the
hydrolysis rate and the Na concentration at which
maximal hydrolysis rate was obtained (Fig. 14A). The
effect of TPP was equivalent to decreasing the
K concentration (Fig. 14B), consistent
with a decreased affinity for K, as demonstrated
above.pNPPase Activity
Lack of effect of TPP on pNPPase activity in the presence of K suggests that the hydrolysis of pNPP does not involve interaction
of K with the extracellular sites. This observation is
in the agreement with the analysis of the kinetics of
K-dependent pNPPase by Beauge and Berberian(1983) and
Berberian and Beauge(1985), who pointed out that ``during
phosphatase activity . . . , the most abundant form is a nonoccluded E and that at least one of the mechanisms of
potassium stimulation of that activity is to take the enzyme into the E state'' (Berberian and Beauge, 1985). due to the
K-stimulated dephosphorylation of EP formed from ATP through the
``physiological route'' (cf. Glynn, 1985). The fact
that under these conditions TPP affected the pNPPase
activity indicates an involvement of the extracellular
potassium-dependent steps of the reaction and supports this hypothesis.
More surprising is that TPP also affected the activity
in the absence of ATP when both NaCl and KCl were present (Fig. 15). Since the transition between EK
and ENa was not affected by the lipophilic cation,
this implies that also in this case the reaction passes through some
intermediates with their outward-facing sites open.Electric Field in the Membrane and Cation Interactions
with Na,K-ATPase: Conclusions
The pronounced effect of
lipophilic ions (Läuger, 1991a) and of the
transmembrane potential on Na and K interaction with the outward-facing sites (Gadsby et
al., 1991; Rakowski, 1991; Rakowski et al., 1991; Schwarz
and Vasilets, 1991; Vasilets et al., 1991; Gadsby et
al., 1992; Stimers et al., 1993; Vasilets and Schwarz,
1993; Hilgemann, 1994; Sagar and Rakowski, 1994) was taken as evidence
that ``extracellular Na and K ions have to pass through a narrow
access channel, which traverses a substantial fraction of the membrane
field, before they can interact with their binding sites on the
Na,K-ATPase molecule'' (Gadsby et al., 1992). The
previously observed weak or even lacking membrane potential dependence
of Na and K interaction with the
inward-facing sites (Rephaeli et al., 1986a; Goldshlegger et al., 1987) and several other observations led to the
conclusion that these sites are formed as shallow, low field, wells
(Goldshlegger et al., 1987; Apell, 1989;
Läuger, 1991, a and b). and K with the extracellular
sites, immersed in the membrane and thus influenced by the
intramembrane electric field, but do not influence the interaction with
the intracellular sites. The electric field-dependent modification of
the interaction of Na,K-ATPase with the extracellular cations was
revealed both in the backward transition of the ADP-insensitive to the
ADP-sensitive EP and in the Na and
K reactivity toward and the kinetic properties of the
K-sensitive EP. The effect of lipophilic ions on the
transition between the phosphoforms is in agreement with the data of
Rephaeli et al. (1986b), who showed a direct influence of the
transmembrane potential on the rate of transition. Our results show
that the rebinding of Na to the ADP-insensitive EP is affected by the intramembrane electric field, but they
do not exclude that the Na deocclusion and/or release
could also be affected. The data indicate that the intramembrane
electric field alters the affinity for extracellular
K, supporting the notion that the electric field
modifies K binding to the outward-facing sites
(Rakowski, 1991; Rakowski et al., 1991; Schwarz and Vasilets,
1991; Stürmer et al., 1991; Vasilets et al., 1991; Stimers et al., 1993; Vasilets and
Schwarz, 1993; Sagar and Rakowski, 1994). Moreover, since the
dephosphorylation rate constant and the ability to occlude
extracellular cations also appear to be altered, our results suggest
some modification of the phosphoenzyme-cation complex itself by the
electric field. or
TPB be caused 1) by lipophilic ion-dependent changes
in surface potential which in turn could influence the access of
cations to their extracellular binding/release sites or 2) by some
charge-dependent changes in protein structure? If the first were true,
a significant modification of lipophilic ion effects by an increasing
ionic strength of the medium, i.e. by ``screening''
of the surface charge, should be expected. We observed some
modification of the interaction of the enzyme with intracellular
Na in measurements of the formation of EP.
The effect of the lipophilic ions on the formation of EP which
we reported previously (Klodos and Plesner, 1992) disappeared when the
ionic strength of the medium was increased. However, neither the
interaction with extracellular Na nor the interaction
with extracellular K appeared to be affected. The
first was shown in ADP chase experiments performed in the presence of
10 mM NaCl without or with 90 mM NMGCl, and the
latter in K chase experiments, where an increase in
KCl concentration from 20 to 50 mM in the K chase i.e. increase in the total salt concentration in
the medium from 120 to 150 mM, did not affect the
K-dephosphorylation. Thus, the lipophilic ion effect
of the cation interaction with the extracellular sites cannot be
ascribed to changes in the surface charge. and TPB have opposite effects speaks against this explanation since it is
difficult to imagine that a modification of different residues would
elicit exactly opposite effects on cation accessibility. However, the
fact that both the dephosphorylation rate constant of the K-sensitive EP and its ability to occlude extracellular cations appear to
be altered indicates some modification of the EP-cation
complex.
), tetraphenylphosphonium;
TPB, tetraphenylboron; Na,K-ATPase, (Na + K)-stimulated adenosine triphosphatase;
K-pNPPase, K-stimulated p-nitrophenylphosphatase; E, Na,K-ATPase
form with high affinity toward ATP and Na; E, Na,K-ATPase form with high affinity toward
K and low affinity to ATP; EP, phosphoenzyme; EP(Na), phosphoenzyme with m occluded Na ions, resistant to K and sensitive to ADP; EP,
ADP-sensitive phosphoenzyme; EP,
phosphoenzyme insensitive to ADP; EP,
phosphoenzyme sensitive to K; EP, phosphoenzyme sensitive to
K; EP, K-insensitive
phosphoenzyme; EP(K),
phosphoenzyme with n occluded K ions; NMG, N-methyl-D-glucamine; MES, 4-morpholineethanesulfonic
acid.
Inge Raungaard and Angielina Tepper are gratefully
acknowledged for their excellent technical assistance. We thank H.-J.
Apell for the gift of lipophilic ions and Mikael Esmann for his help
and advice in carrying out the experiments with eosin and for helpful
discussions. We thank Sergey N. Fedosov, Jens G. Nørby, Robert
L. Post, and Igor W. Plesner for many stimulating discussions,
encouragement, and help in writing the manuscript.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Starke-Peterkovic, N. Turner, P. L. Else, and R. J. Clarke Electric field strength of membrane lipids from vertebrate species: membrane lipid composition and Na+-K+-ATPase molecular activity Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2005; 288(3): R663 - R670. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |