| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Child Health, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, United Kingdom
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
We describe how cations, in the presence of ATP, regulate the
phosphorylated form of 19- and 21-kDa nucleoside diphosphate kinase
(NDPK; EC 2.7.4.6), a kinase controlling
K+ channels, G proteins, cell
secretion, cellular energy production, and UTP synthesis. In apically
enriched human nasal epithelial membranes, 10 mM
Na+ inhibits phosphorylation of
NDPK relative to other cations. Dose response showed that, whereas
K+ induces a fourfold greater
phosphate incorporation (EC50 10 mM), Na+ is inhibitory
(EC50 10 mM) compared with
respective buffer controls. Cation discrimination is nucleotide
selective (not seen with
[
-32P]GTP) and NDPK
specific (not seen with p37h, a previously characterized Cl
-sensitive
phosphoprotein). Na+ does not
exert an inhibitory effect on NDPK phosphorylation directly but is
likely to act via an okadaic acid-insensitive phosphatase. We speculate
that the ability of NDPK to discriminate between physiologically
relevant cation concentrations provides a novel example of cross talk
within the apical membrane.
membrane; phosphatase; nucleotide; adenosine 5'-triphosphate; chloride
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
THE REGULATION OF Na+ absorption from the apical surface of epithelial cells involves the coordination of the following processes: Na+ entry to the subapical cytoplasm via Na+ channels, consumption of energy for maintenance of a steady intracellular Na+ concentration ([Na+]i) via activation of the basolaterally located Na+-K+-ATPase, and prevention of the rise in intracellular K+ concentration ([K+]i) by extrusion through basolateral K+ channels (17, 32). Neither the process (which involves cross talk) nor the molecular linkages are understood. Here we describe the properties of a prerequisite for a coherent model: a kinase differentially regulated by [Na+]i and [K+]i.
There are at least three (nonexclusive) ways in which Na+ may control its own transepithelial flux. These include 1) inhibition of Na+ entry ("self-inhibition") by extracellular Na+ concentration ([Na+]o; see review in Ref. 32), 2) regulation of Na+ channel recruitment from subapical pools (34), and 3) [Na+]i acting on an intracellular signaling pathway, a notion we expand in this paper. Thus Komwatana et al. (15) described saturation of Na+ conductance as the [Na+]o increased in mouse mandibular duct cells, and Turnheim (32) has reviewed the evidence for an external Na+ sensor. Van Driessche and Lindemann (34) found that [Na+]o-dependent saturation of Na+ transport was due to a decline in Na+ channel density, a process that Els and Chou (7) found to be sensitive to cytochalasin B in frog skin. However, the simple interpretation that the density of Na+ channels, and therefore the rate of Na+ transport, relied on (microfilament-dependent) recruitment of new channel proteins has been challenged by recent studies (1). The new data suggest that cytochalasin B may inhibit the translocation of an inhibitory protein kinase to the channel. The authors of these studies showed (using Xenopus oocytes and planar lipid bilayers) that activity of the amiloride-blockable epithelial Na+ channel was inhibited by protein kinase C (PKC) but found that inhibition was 80% attenuated by pretreatment with cytochalasin B. These experiments introduced the idea that PKC-dependent phosphorylation is involved in the inhibition of Na+ channel activity, but it should be noted that, in this system, protein kinase A did not activate the channel. The third model speculates that Na+ interacts with second messengers. This model predicts an intracellular sensor(s) for Na+ within a signaling cascade(s), an idea we explore here by defining the relationship between the degree of phosphorylation of a kinase [nucleoside diphosphate kinase (NDPK)] and the cation species bathing apical membrane fractions derived from the Na+-absorptive epithelium of the human upper airway.
The current study complements our earlier demonstration of
a phosphorylation-mediated,
Cl concentration
([Cl])-dependent
sensor cascade in human (30, 31) and sheep airway epithelium (21). We
reported that the rate of change in the in vitro phosphorylation
profile of a 37-kDa protein (p37h or p37s; suffixes denote human and
sheep, respectively) was maximal on either side of the steady-state
[Cl]i
in airway epithelial cells (~40 mM; see Ref. 30). We speculated that
the role of the membrane-bound kinase(s) and or phosphatase(s) regulating the net phosphorylation of p37h or p37s could be to "feed
back"
[Cl]i
to membrane transporters and/or channels, which would explain the well-established connection between
[Cl]i
and the membrane conductance of
Na+ and
Cl (5, 6, 18, 20, 29, 36).
During the course of these studies, we found that
Na+, in a concentration range
expected for the epithelial cytoplasm (~10 mM), inhibited apical
membrane phosphorylation (Mehta, unpublished observations). The present
study had two aims, 1) to study the effects of cations on apical membrane phosphorylation by clamping [Cl] below the
concentration that affects phosphorylation and
2) to show that
Cl interacts with
Na+ and
K+ to exert, respectively,
inhibitory and permissive effects on NDPK phosphorylation. We observe
that NDPK (but not p37h) is able to discriminate between
K+ and
Na+ at physiologically relevant
intracellular concentrations for each cation. We also show that, unlike
our previously reported Cl-dependent cascade, which
could use either GTP or ATP as kinase substrates,
Na+ and
K+ exert their differential
regulatory effects on NDPK when ATP is the phosphate donor.
| |
MATERIALS AND METHODS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Collection of Human Respiratory Epithelial Cells
The subjects studied were all healthy young adults under the age of 45 yr undergoing surgery for reasons unrelated to nasal mucosal disease. Written informed consent and Tayside Committee on Medical Research Ethics approval were obtained. As detailed in Treharne et al. (31), immediately after anesthesia, respiratory epithelial cells were brushed from the inferior nasal turbinate epithelium with a cytology brush and dislodged into a nutrient medium (medium 199). Apically enriched membranes were then prepared. Briefly, a disrupted postnuclear supernatant was fractionated on a discontinuous sucrose gradient, and apical membranes were identified by alkaline phosphatase enrichment. Contaminating membranes were excluded by appropriate marker assays.Phosphorylation of Membranes by Endogenous Kinase(s)
Apically enriched membrane pellets (made from pooled brushings, typically from 4 individuals) were resuspended in ice-cold 10 mM MOPS (pH 7.9 with KOH, final concentration of K+ 11 mM) containing 0.05% Triton X-100 (referred to as "membrane buffer"), with 20 µM dithiothreitol and 0.5% DMSO. Apical membrane proteins were incubated with various 10 mM cation-Cl combinations (20 min at 4°C), followed by the addition of
[-32P]ATP for 5 min
at 37°C. For the cation control experiments, MOPS buffer (pH 7.9)
was made using either tetramethylammonium (TMA) hydroxide (final
concentration 13 mM) or a mixture of KOH and NaOH (final concentrations
of 4 and 13 mM, respectively). For the dose response studies
(0-100 mM
cation-Cl), 10-µg
aliquots of apically enriched membranes were incubated with 37 kBq of
either [-32P]ATP or
[-32P]GTP (final
concentration 7 nM) at 37 or 4°C for the appropriate time and
conditions. In all experiments, 1 µl of
[-32P]ATP or
[-32P]GTP (within 1 half-life of activity date) was spotted onto the side of the tube, and
the reaction was started by a rapid spin to mix reagents.
Phosphorylation was terminated by addition of 5× Laemmli sample
buffer (16), followed by rapid mixing.
Quantitation of Phosphorylation
Proteins were separated by SDS-PAGE using 12.5% polyacrylamide gels on a Protean II slab cell (Bio-Rad). Prestained molecular mass markers were used to avoid the loss of phosphohistidine in the acid environment of staining and destaining of gels before quantification. The incorporation of phosphate into individual protein bands was quantified using electronic autoradiography (Canberra-Packard Instant Imager).Method 1: normalizing against the cation inducing maximal phosphorylation. Method 1 is described in Ref. 30. Briefly, ordinate data (see Fig. 1) are means ± SE of percent net maxima. The percent net maximum for a given protein is calculated to compensate for the variation in intensity of phosphorylation between pooled membrane samples from different individuals. Data were normalized to the maximally phosphorylated band for each experiment, minus background. Background phosphorylation was defined as an area of the gel adjacent to the protein of interest but containing no phosphorylated proteins. Typically, apical membranes were bathed in a variety of cations. Thus, for a given protein (e.g., p19h), when the intensity of its phosphorylation was measured, the p19 lane with the greatest incorporated counts/min (cpm) was first assigned a value of 100%, and the rank order of the remainder were expressed relative to this value (%net maximum).
Method 2: normalizing phosphorylation relative to buffer control. During the dose-response studies, the object was to compare relative phosphate incorporation in the presence of Na+, K+, or NH+4. Thus all cpm incorporated into a given protein in the presence of buffer control were given the arbitrary value of 1 and the other cations were compared against this reference.
Immunoprecipitation of NDPK
Immunoprecipitation of NDPK has been described recently (21).Chemical Reagents
All chemicals were of analytical grade and were purchased from Sigma or BDH (Poole, UK) except for the following: [-32P]ATP and
[-32P]GTP were from
NEN DuPont (Stevenage, UK), the acrylamide and other electrophoresis
materials were from Bio-Rad, PBS was from Microgen Bioproducts, okadaic
acid was from Calbiochem (Nottingham, UK), medium 199 was from Flow
Laboratories, and NDPK antibodies were from Santa Cruz Biotechnology.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Cations Regulate Phosphorylation of p19h and p37h
Figure 1 shows the results of bathing apically enriched human airway membranes in one of the Cl salts of eight different
cations (10 mM; Na+,
K+,
NH+4,
Rb+,
Li+, choline, tetraethylammonium,
or TMA) before the initiation of phosphorylation of apical membrane
proteins. The electronic autoradiograph (Fig.
1A) shows that the intensity of
phosphorylation of a pair of 19- and 21-kDa proteins (p19h and p21h)
and p37h is differentially dependent on the cation species bathing the
membrane. We have recently shown that in sheep tracheal epithelium,
p19s and p21s are isoforms of NDPK located within apically enriched
membranes (21). We found that p21s was more sensitive than p19s to
dephosphorylation by Mg2+, thus
making it difficult to study because trace amounts of
Mg2+ are essential for kinase
activity. In the present study, p21h phosphorylation was so faint that
it could not be accurately discriminated against background. Thus p19h
phosphorylation was quantitated and showed significantly enhanced
phosphorylation with Rb+, choline,
and NH+4 (Fig.
1C) compared with 10 mM
Na+. A similar cation-dependent
pattern was seen for p37h, which was maximally phosphorylated with
Rb+ (Fig.
1B). The two key differences between
these phosphoproteins were that for p37h the presence of
Na+ resulted in the least
phosphorylation compared with all the remaining cations and, second,
that p37h was more intensely labeled than p19h.
|
Identification of p21h as NDPK
Recently, we have shown that apical membranes derived from sheep airway provide a good model for Cl-dependent human airway
phosphorylation (21). In that study, we used immunoprecipitation to
identify p21s as an isoform of NDPK. The native form of NDPK is a
hexameric protein made up of different combinations of 19- and 21-kDa
proteins (37). Commercial antibodies are only available for the larger
21-kDa isoform (known in an alternative nomenclature as nm23-H1).
Qualitatively, p21h phosphorylation mirrored that of p19h (see Figs. 1
and 3-5; see also Ref. 21). We therefore tested the idea that p21h
was human nm23-H1 NDPK using polyclonal antibodies raised against the
nm23-H1 human NDPK. Figure
2A shows
the starting material, i.e., the phosphoproteins within a 20-µg
aliquot from the phosphorylated membranes used for the
immunoprecipitation. Figure 2B shows
that a single phosphorylated band of 21 kDa was present in an aliquot
from the final immunoprecipitate. Prior incubation of the anti-nm23-H1
antibody with the peptide against which it was raised eliminated the
immunoprecipitation of this 21-kDa phosphoprotein (Fig.
2C). In further controls (data not
shown), we could not immunoprecipitate any phosphorylated proteins
using an unrelated antibody to the common antigen of the 
-subunits
from G proteins. Having identified the 21-kDa isoform of NDPK, we
increased the concentration of both the membranes and the antibody
(10-fold) and were able to immunoprecipitate a phosphoprotein of 19 kDa
in addition to p21 (Fig. 2D). A
similar doublet was immunoprecipitated from a postnuclear supernatant of the H441 airway cell line (data not shown). These data, together with the evidence in Muimo et al. (21), suggest that NDPK is present in
our phosphorylation cascade. We investigated the effects of
Na+ and
K+ concentration on the
phosphorylation of the p19h-NDPK.
|
Na+ and K+ Have Opposite Effects on p19h-NDPK Phosphorylation
Comparison of the electronic autoradiographs in Figs. 3A and 4A shows that the phosphorylation of p19h-NDPK declines as NaCl concentration ([NaCl]) increases (Fig. 3), whereas the reverse is seen when [KCl] increases (Fig. 4). Two quantitative aspects of the cation dependence of the phosphorylation of p19h-NDPK are described (Figs. 3C and 4C). First, a quasi-exponential, [NaCl]-dependent decline in phosphorylation is present (EC50 10 mM). Second, the pattern for [KCl] shows the exact opposite: a concomitant increase in phosphorylation of p19h-NDPK (EC50 10 mM for added K+; see Buffer Controls and Inhibitory Effects of Na+). It is noteworthy that as phosphorylation reached its nadir (by increase of [NaCl] from 0 to 25 mM), p19h-NDPK phosphorylation approached its maximum with the addition of 25 mM [KCl]. The decline in phosphorylation induced by NaCl was a Na+-specific effect, because it could not be reproduced with NH4Cl, the latter mimicking K+ and showing an increase in the phosphorylation of p19h-NDPK (Fig. 4C).
|
|
Na+ and K+ Have Similar Effects on p37h Phosphorylation
Figure 3B shows that 50 mM NaCl induces a fivefold reduction in p37h phosphorylation compared with buffer control. Because the results for p19h-NDPK suggested that K+ and Na+ acted reciprocally, we used p37h phosphorylation to test the generality (or otherwise) of our earlier observations. The gel in Fig. 4A shows that increasing [KCl] or [NH4Cl] results in a decline in p37h phosphorylation irrespective of the cation-Cl combination. This
indicates that p37h phosphorylation is regulated differently from
p19h-NDPK phosphorylation. Interestingly, the pattern of decline is
also different (compare Figs. 3B and
4B). As [KCl] increases,
there is a quasi-linear decline in p37h phosphorylation (EC50 25 mM) compared with the
exponential fall seen with NaCl (EC50 10 mM; Fig. 3).
KCl-dependent phosphorylation of p37h continues its linear descent
beyond 50 mM, whereas for NaCl there was no significant decline after
50 mM.
Buffer Controls and Inhibitory Effects of Na+
The MOPS acid buffers were neutralized with KOH before the addition of membranes and therefore already contained an initial concentration of 11 mM K+ before the 10 mM cation-Cl combinations were
added. It remained possible that this excess of
K+ was falsely enhancing p37h and
p19h phosphorylation relative to
Na+. However, replacement of the
K+ with TMA or a combination of
KOH and NaOH had no effect on the Na+-dependent decline in p37h and
p19h phosphorylation (Fig. 5,
A and
B). Once again, the presence of
Na+ resulted in the least membrane
protein phosphorylation. We have reported previously (30) that the
effects of increasing salt concentration were not due to the
colligative properties of the phosphorylation buffer, because they were
not reproduced with mannitol or sucrose. We concluded that
Na+ was an important inhibitory
influence on both p37h and p19h-NDPK phosphorylation, whereas
K+ (and other cations) were
stimulatory to NDPK alone.
|
Low Temperature Eliminates Na+ Inhibition of NDPK Phosphorylation
The cation-dependent changes in phosphorylation could have been mediated via kinase(s), phosphatase(s), or both. All the results described so far relate to steady-state experiments (phosphorylation for 5 min at 37°C) in which both sets of enzymes were (potentially) active. At 4°C kinases have been reported to retain their activity relative to phosphatases (3), providing a simple method of discriminating between their relative contributions. To define which reaction conferred generation of cation sensitivity, we repeated the NaCl and KCl dose responses at the lower temperature. Figure 6 shows the typical effects of increasing [NaCl] or [KCl] at 4°C and reveals five interesting results: 1) p37h is no longer phosphorylated; 2) p19h and p21h are the sole phosphorylated species, probably reflecting autophosphorylation of NDPK (21); 3) Na+ and K+ now have similar effects on the phosphorylation of p19h-NDPK (compare with Figs. 3 and 4); 4) lowering the reaction temperature generates a new profile of salt-dependent phosphorylation that differs from that seen at 37°C with either cation-Cl combination: at
4°C, p19h phosphorylation declines with either cation (contrast
result at 37°C; 25 mM NaCl was selectively inhibitory; see Fig. 3);
and 5) phosphorylated p21h is
measurable for the first time and mirrors the profile of p19h, albeit
at a lower intensity (quantitative data not shown). We conclude that at
4°C discrimination between
Na+- and
K+-dependent phosphorylation is
absent. This discrimination in phosphorylation at 37°C with
Na+ suggests the presence of a
Na+-sensitive phosphatase(s).
|
Role of Phosphatases
Our previous data showed that the kinases in this system were insensitive to conventional inhibitors (30), and we implicated phosphatase involvement because of the loss of Cl-dependent
phosphorylation when the phosphatase-resistant analog adenosine
5'-O-(3-thiotriphosphate) was
substituted for ATP (31). The conventional approach to understanding
the role of phosphatases is to add specific inhibitors of known
phosphatases. This approach was not successful because no differences
in phosphorylation were observed when experiments were conducted in the
presence or absence of okadaic acid (Fig.
7). Microcystin LR and calyculin A were equally ineffective (data not shown). To show that the relationship between the putative phosphatase(s) and
Na+ was not due to an anion, we
incubated membranes in okadaic acid (1 µM) and then added two
different Na+-anion combinations:
NaCl or sodium gluconate. We used 40 mM
Na+ (fourfold above
EC50) and found that,
irrespective of the accompanying anion, okadaic acid failed to restore
(Na+-inhibited) phosphorylation of
p19h (Fig. 7), confirming our earlier observations that
Na+ was not activating protein
phosphatase 1 or 2A (PP-1 or PP-2A) to induce a decline in p19h
phosphorylation. This experiment also revealed an unexpected result
with respect to the species of anion accompanying the
Na+. In the presence of sodium
gluconate, p19h phosphorylation was significantly elevated compared
with NaCl, suggesting that 1) the
species of anion influences p19h dephosphorylation and
2) a synergy exists between
Na+ and
Cl in the dephosphorylation
of NDPK. The latter idea has resonance with our earlier data (see Fig.
2 of Ref. 30) showing that gluconate and
Cl lie at opposite ends of
the phosphorylation spectrum when other phosphoproteins within this
apical cascade of phosphorylation are studied. Although these data are
consistent with a phosphatase-mediated Na+-activated dephosphorylation of
NDPK, Na+-dependent inhibition of
the kinase could still have occurred. We excluded this possibility by
measuring initial (quasi-zero time) incorporation of phosphate into
NDPK in the presence or absence of
Na+. Membranes were incubated with
[-32P]ATP on ice
for 10 s in 10 mM NaCl. We found that the cpm incorporated into NDPK in
the presence of Na+ exceeded
buffer control by an average factor of 2 (mean ± range 651 ± 153 vs. 281 ± 113 cpm in the presence and absence of
Na+, respectively), thus excluding
the notion of an inhibitory
Na+-kinase interaction.
|
Nucleotide Substitution Eliminates Na+ Inhibition of Phosphorylation
Our previous results had shown that the Cl sensitivity of p37h
phosphorylation occurred principally but not exclusively (depending on
the cation-anion combination) when
[-32P]GTP was the
kinase substrate. We therefore tested the effects of this guanine
nucleotide on the above profile of phosphorylation. Apical membrane
proteins were phosphorylated for 5 min at 37°C with
[-32P]GTP instead
of [-32P]ATP in the
presence of increasing concentrations of NaCl, KCl, or
NH4Cl. Quantitation of p37h
phosphorylation from ATP and GTP is shown in Fig.
8, A and
B, respectively: in the presence of [-32P]GTP, NaCl
induces a quasi-linear decline in phosphorylation of p37h. We observe
that p37h phosphorylation now decreases in a similar manner,
irrespective of the cation species (Fig. 8; compare
A and
B). In addition, for p19h-NDPK
phosphorylation from [-32P]GTP,
Na+ now has two different effects.
First, no inhibition of phosphorylation occurs, and, second, a slow
increase in phosphorylation occurs as NaCl approaches 100 mM (probably
Cl mediated; see Fig. 1 of
Ref. 21).
|
The combined evidence suggests that the
Na+-dependent regulation of
phosphorylation, in a range that is likely to be present in the
intracellular environment (0-25 mM
Na+), only occurs when ATP is
present as the kinase substrate. This is further illustrated by
quantitating the phosphorylation of p19h with
[-32P]ATP (Fig.
8C). The phosphate incorporation
into p19h-NDPK shows (relative to buffer control) a fourfold excess in
the presence of K+. In contrast,
there is a 50% decline in phosphorylation with Na+.
NH+4 occupies an intermediate position.
However, in the presence of
[-32P]GTP,
p19h-NDPK cannot discriminate between cation species. We also find that
p37h is inhibited by either cation
(Na+ > K+), suggesting
Na+-dependent dephosphorylation is
specific to NDPK.
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Control of Vectorial Ion Transport in Epithelia
Ion substitution is a general stratagem for the study of ion transport across cell membranes. Our study suggests that ion substitution is not a neutral process where the Na+-absorptive apical membrane of airway epithelium is concerned. Apical membranes from this epithelium experience a high-K+ cytosolic environment and yet regulate the local rise in Na+ concentration during Na+ absorption while maintaining a low [Na+]i. This membrane displays a complex interaction (Refs. 21 and 30 and this study) between ion species and concentration, nucleotides, and the phosphorylation state of NDPK, a kinase known to regulate K+ channels (10, 38), secretion (19), and cellular energy metabolism (35). We now show that membrane-bound NDPK discriminates between K+ and Na+ at intracellular concentrations physiologically relevant for each cation in the presence of ATP but not GTP. This is contrary to the literature, which reports that purified NDPK does not discriminate between nucleotides (22, 23, 37). We show that Na+ reduces the net phosphorylation of membrane-bound NDPK in the presence of ATP alone. This difference in phosphorylation between ATP and GTP may be explained either by NDPK being phosphorylated on different sites with each nucleotide or by a GTP-mediated inhibition of the phosphatase(s) targeting the kinase.Consequences of Ion Regulation of NDPK
In epithelia, it is well recognized that the opening of basolaterally located K+ channels provides the driving force for secretion via the apical membrane (17). At the apical membrane, UTP is an important Cl secretagogue (11, 29),
and phosphorylated NDPK provides the sole pathway for UTP synthesis
(from UDP and ATP). We now find that the anion
(Cl 
gluconate)
accompanying Na+ modulates the
degree of inhibition of NDPK phosphorylation. Because we know that
a
[Cl]i
of >40 mM enhances NDPK phosphorylation (21), our results provide a
plausible feedback loop between UTP synthesis and
[Na+]i
and
[Cl]i.
Specifically, as
[Na+]i
rises, NDPK phosphorylation falls, but in the presence of a Na+ substitute
[N-methyl-D-glucammonium
(NMDG), K+, or
NH+4] NDPK phosphorylation does not
decline. Our data predict that a high
[Na+]i
will be inhibitory to UTP-mediated processes in human airway because
Na+ overcomes the enhancing effect
of KCl on NDPK phosphorylation.
Multiple Functions of NDPK in Other Systems
By an unknown mechanism, NDPK regulates ACh-activated K+ channels (KACh channels) (38). When a muscarinic agonist is present, KACh channels open, and antibodies to NDPK inhibited this opening. However, simultaneous suppression of GTP production via NDPK does not prevent kinase-mediated channel inhibition. This dissociation between the nucleotide synthetic function of NDPK and its role as a conventional kinase is well recognized but poorly understood (9, 19, 35). The same study (38) also found that inhibition of NDPK had no effect on baseline KACh channel activity in the absence of agonist, suggesting that channel modulation by NDPK is a "poststimulus" event. It is precisely under these conditions that intracellular ion concentrations change rapidly. The recognized functions of this kinase, regulation of pancreatic secretion and cellular energy metabolism, are compatible with an integrative role for NDPK in membrane channel function via ions as second messengers. Our current and previous data with anions (21, 30) point to a phospho-relay, which is ideally suited to respond to ionic perturbations in epithelia. NDPK is ideally positioned to sense ions on the cell surface (33), on the inner leaflet of the plasma membrane (12), and in the cytoplasm (22). The concept of an ion sensor is particularly relevant to human nasal airway epithelium, which is subject to a wide variety of environmental challenges (temperature, heat, relative humidity, and so forth) with a potential to disturb the tonicity and/or composition of the airway surface liquid bathing the cilia and thereby the composition of the epithelial cytosol (17, 30). Despite these potential perturbations, [Na+]i is held constant at ~10-fold below its plasma value (25, 26), precisely in the range of maximum Na+ sensitivity of NDPK phosphorylation.Na+ is unlikely to inhibit NDPK phosphorylation directly, because our "initial rate experiments" show that phosphate incorporation into the kinase is not reduced in the presence of 10 mM NaCl. Indeed, the observed twofold elevation of phosphate incorporation with Na+ over buffer control is consistent with our proposal that a Na+-activated phosphatase is present. This novel notion is supported by the loss of Na+ sensitivity at 4°C, a temperature reported to inhibit phosphatase activity (Ref. 3, see also Ref. 28). Although the phosphatase is unknown, our data suggest that it cannot be PP-1 or PP-2A (okadaic acid has no effect) and is unlikely to be PP-2B or PP-2C (ovine airway data, Ref. 21). A 10-kDa Na+-dependent phosphatase from bacteria with maximal in vitro activity directed against human NDPK has been described (27). This phosphatase has very little activity in the presence of K+, compatible with our data.
Cation-Anion Interactions With Apical Membrane Phosphoproteins
The Na+-dependent inhibition of NDPK phosphorylation was less marked when the accompanying anion was switched from Cl to
gluconate. This anion dependence is reminiscent of our previously characterized Cl-dependent
protein p37h, whose phosphorylation was enhanced by gluconate compared
with Cl, albeit in the
presence of GTP (and not ATP; note the contrast with the current
results). The combined data suggest that anion and cation
differentially regulate NDPK and p37h phosphorylation using different
nucleotides. However, we do not understand the differences in the
pattern of Cl dependence
between NDPK and p37h. In the present study, p37h phosphorylation shows
no difference in sensitivity between
Na+ and
K+ with either ATP or GTP.
Nevertheless, the anion-cation combination is important for the net
phosphorylation of p37h. For example, a dose response with NMDG
chloride induces a concentration-dependent peak of phosphorylation
(~40 mM), whereas NaCl induces an exponential decline and KCl induces
a quasi-linear decline.
Although the NDPK phosphorylation reported in this paper is likely to be autophosphorylation, it remains possible that another Na+-sensitive kinase phosphorylates NDPK. Casein kinase 2 (CK2)-dependent phosphorylation of serine adjacent to the phosphohistidine site in NDPK is known to prevent NDPK autophosphorylation (2, 8). That 100 mM NaCl promotes CK2 activity in vitro is consistent with the second kinase hypothesis. Consequently, in this model, Na+ could activate CK2 to inhibit NDPK phosphorylation indirectly. However, such a model could not readily explain why a reduction in temperature eliminates Na+-dependent inhibition, since kinase activity is not likely to be curtailed. We are not aware of data on the temperature dependence of CK2 activity.
NDPK as an Integrator of
[Na+]i,
[K+]i,
and
[Cl]i
]i
and Cl conductance through
the cystic fibrosis transmembrane conductance regulator (36). Thus
Cl regulates
Cl exit. Furthermore,
[Cl]i
regulates Cl entry via the
Na+-K+-2Cl
cotransporter (18) and additionally controls cation transport (6).
Dinudom et al. (6) showed that as
[Cl]i
increased, the inward Na+ current
decreased. Simultaneously, the inward
Cl current increased. In
rat fetal distal lung epithelium, Marunaka et al. (20), studying
Na+ absorption, found that
activation of a nonspecific cation channel depended on
[Cl]i.
Thus there appears to be a relationship between the identity and
concentrations of intracellular ions and their rates of ion transport,
but the links are poorly understood. Dinudom et al. (5) reported that
the
[Cl]i
regulates Na+ transport via a G
protein. However, further complexity is introduced by the known
Cl sensitivity of some G
proteins (see Ref. 21 for discussion). NDPK associates with G proteins,
but the functional consequences of this binding are controversial. It
has been suggested that NDPK and
Gs may coexist such that the
complex could allow the localized formation and exchange of GTP for GDP
bound to the G protein (13). Randazzo et al. (24) rebutted this concept
but used a model system that did not control for the presence of
Na+ (their buffers were based on
Na+-neutralized HEPES).
Interestingly, it was the use of such buffers that led us to the
preliminary observations on the inhibitory effects of
Na+ in our system. The above
authors concluded that, since they could find no evidence for a role of
NDPK in GDP-to-GTP conversion when the GDP was bound to G proteins, the
role of NDPK as a local GTP generator was untenable. Our data suggest
that changes to the ionic composition of the buffer may alter these
results. Alternatively, the kinase function of NDPK could operate
despite inhibition of its phosphotransferase role and regulate G
proteins by a different mechanism (see above and Ref. 9).
The airway epithelium both secretes and absorbs fluid, but it is
unclear whether or not both processes occur in a single cell within the
same time frame. Our work points to a molecular switch between these
processes regulated by ion-dependent phosphorylation. We propose the
following model (Fig. 9). When
[Cl]i
is high and
[Na+]i
is <10 mM, NDPK is phosphorylated (on one or more sites). Assuming this phospho-NDPK is essential for function (UTP synthesis and K+ channel opening, for example),
we predict that as K+ leaves the
cell, the resultant membrane hyperpolarization "coupled" to the
synthesis of UTP will drive apical
Cl secretion. However, the
mechanism that might transport this UTP to the airway lumen is unclear.
The resultant fall in
[Cl]i
will turn off the kinase as well as promote KCl entry via the Na+-K+-2Cl
cotransporter, completing the feedback loop. Should cAMP-activated apical Na+ absorption be
occurring, the membrane-delimited rise in
[Na+]i
will itself turn off NDPK via phosphatase activity. The latter predicts
that a high
[Na+]i
turns off UTP production. Finally, our unusual finding with respect to
the anion accompanying the Na+
suggests a two-site model for ion binding within this
cascade.
|
| |
ACKNOWLEDGEMENTS |
|---|
We thank Anna Crichton for collecting nasal brushings and the Ninewells theater staff for help over many years.
| |
FOOTNOTES |
|---|
L. J. Marshall is a Wellcome Prize Student, R. Muimo was supported by the Wellcome Trust, and C. E. Riemen was supported by the United Kingdom Cystic Fibrosis Trust. BioMed II Grant BMH4-CT96-0602 and Tenovus (Scotland) provided support and purchased equipment; the Anonymous Trust provided ongoing support.
Address for reprint requests: A. Mehta, Dept. of Child Health, Ninewells Hospital Medical School, University of Dundee, Dundee DD1 9SY, UK.
Received 3 December 1997; accepted in final form 30 September 1998.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
1.
Awayda, M. S.,
I. I. Ismailov,
B. K. Berdiev,
C. M. Fuller,
and
D. J. Benos.
Protein kinase regulation of a cloned epithelial sodium channel.
J. Gen. Physiol.
108:
49-65,
1996
2.
Biondi, R. M.,
M. Engel,
M. Sauane,
C. Welter,
O. Issinger,
L. J. de Asua,
and
S. Passeron.
Inhibition of NDPK activity by in vitro phosphorylation of protein kinase CK2.
FEBS Lett.
399:
183-187,
1996[Medline].
3.
Brautigan, D. L.,
and
C. L. Shriner.
Methods to distinguish various types of protein phosphatase activity.
Methods Enzymol.
159:
339-346,
1988[Medline].
4.
De La Rosa, A.,
R. L. Williams,
and
P. S. Steeg.
Nm23/nucleoside diphosphate kinase: toward a structural and biochemical understanding of its biological functions.
Bioessays
17:
53-62,
1995[Medline].
5.
Dinudom, A.,
P. Komwatana,
D. I. Young,
and
D. I. Cook.
Control of the amiloride-sensitive Na+ current in mouse salivary ducts by intracellular anions is mediated by a G protein.
J. Physiol. (Lond.)
487:
549-555,
1996[Medline].
6.
Dinudom, A.,
D. I. Young,
and
D. I. Cook.
Sodium and chloride conductances are controlled by cytosolic chloride concentration in the intralobular duct cells of mouse mandibular glands.
J. Membr. Biol.
135:
289-295,
1993[Medline].
7.
Els, W. J.,
and
K. Chou.
Sodium-dependent regulation of epithelial sodium channel densities in frog skin; a role for the cytoskeleton.
J. Physiol. (Lond.)
462:
447-464,
1993
8.
Engel, M.,
O.-G. Issinger,
I. Lascu,
T. Seib,
S. Dooley,
K. D. Zang,
and
C. Welter.
Phosphorylation of nm23/nucleoside diphosphate kinase by casein kinase 2 in vitro.
Biochem. Biophys. Res. Commun.
199:
1041-1048,
1994[Medline].
9.
Engel, M.,
M. Véron,
B. Theisinger,
M.-L. Lacombe,
T. Seib,
S. Dooley,
and
C. Welter.
A novel serine/threonine-specific protein phosphotransferase activity of Nm23/nucleoside-diphosphate kinase.
Eur. J. Biochem.
234:
200-207,
1995[Medline].
10.
Heidbuchel, H.,
G. Callewaert,
J. Vereecke,
and
E. Carmeliet.
Acetylcholine-mediated K+ channel activity in guinea-pig atrial cells is supported by nucleoside diphosphate kinase.
Pflügers Arch.
422:
316-324,
1993[Medline].
11.
Hwang, T.-H.,
E. M. Schwiebert,
and
W. B. Guggino.
Apical and basolateral ATP stimulates tracheal epithelial chloride secretion via multiple purinergic receptors.
Am. J. Physiol.
270 (Cell Physiol. 39):
C1611-C1623,
1996
12.
Kimura, N.,
and
N. Shimada.
Membrane-associated nucleoside diphosphate kinase from rat liver. Purification, characterization, and comparison with cytosolic enzyme.
J. Biol. Chem.
263:
4647-4653,
1988
13.
Kimura, N.,
and
N. Shimada.
Evidence for complex formation between GTP binding protein (Gs) and membrane-associated nucleoside diphosphate kinase.
Biochem. Biophys. Res. Commun.
168:
99-106,
1990[Medline].
14.
Komwatana, P.,
A. Dinudom,
J. A. Young,
and
D. I. Cook.
Cytosolic Na+ controls an epithelial Na+ channel via the GO guanine nucleotide-binding regulatory protein.
Proc. Natl. Acad. Sci. USA
93:
8107-8111,
1996
15.
Komwatana, P.,
A. Dinudom,
J. A. Young,
and
D. I. Cook.
Control of the amiloride-sensitive Na+ current in salivary duct cells by extracellular sodium.
J. Membr. Biol.
150:
133-141,
1996[Medline].
16.
Laemmli, U. K.
Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature
227:
680-685,
1970[Medline].
17.
Liedtke, C. M.
Electrolyte transport in the epithelium of pulmonary segments of normal and cystic fibrosis lung.
FASEB J.
6:
3076-3084,
1992[Abstract].
18.
Lytle, C.,
and
B. Forbush III.
Regulatory phosphorylation of the secretory Na-K-Cl cotransporter: modulation by cytoplasmic Cl.
Am. J. Physiol.
270 (Cell Physiol. 39):
C437-C448,
1996
19.
Martin, M. W.,
A. J. O'Sullivan,
and
B. D. Gomperts.
Inhibition by cromoglycate and some flavonoids of nucleoside diphosphate kinase and of exocytosis from permeabilized mast cells.
Br. J. Pharmacol.
115:
1080-1086,
1995[Medline].
20.
Marunaka, Y., T. Narkahari, and H. Tohda. Cytosolic
[Cl] regulates
Na+ absorption in fetal alveolar
epithelium?: roles of cAMP and
Cl channels.
Jpn. J. Physiol. 44, Suppl. 2: S281-S288, 1994.
21.
Muimo, R.,
S. Banner,
L. J. Marshall,
and
A. Mehta.
Nucleoside diphosphate kinase and Cl-sensitive protein phosphorylation in apical membranes from ovine airway epithelium.
Am. J. Respir. Cell Mol. Biol.
18:
270-278,
1998
22.
Parks, R. E., Jr.,
and
R. P. Agarwal.
Nucleoside diphosphokinases.
In: The Enzymes, edited by P. D. Boyer. New York: Academic, 1973, p. 307-331.
23.
Prescan, E.,
A. Vonica,
and
I. Lascu.
Nucleoside diphosphate kinase from human erythrocytes: purification, molecular mass and subunit structure.
Fed. Eur. Biochem. Soc.
250:
629-632,
1989.
24.
Randazzo, P. A.,
J. K. Northup,
and
R. A. Kahn.
Regulatory GTP-binding proteins (ADP-ribosylating factor, Gt and RAS) are not activated by NDPK.
J. Biol. Chem.
267:
18182-18189,
1992
25.
Robertson, M. A.,
and
K. J. Foskett.
Na+ transport pathways in secretory acinar cells: membrane cross talk mediated by [Cl]i.
Am. J. Physiol.
267 (Cell Physiol. 36):
C146-C156,
1994
26.
Sarkadi, B.,
and
J. C. Parker.
Activation of ion transport pathways by change in cell volume.
Biochim. Biophys. Acta
1071:
407-427,
1991[Medline].
27.
Shankar, S.,
A. Kavanaughy-Black,
S. Kamath,
and
A. M. Chakrabarty.
Characterization of a phosphoprotein phosphatase for the phosphorylated form of nucleoside-diphosphate kinase from Pseudomonas aeruginosa.
J. Biol. Chem.
270:
28246-28250,
1995
28.
Somero, G. N.
Temperature and proteins: little things can mean a lot.
News Physiol. Sci.
11:
72-77,
1996.
29.
Stutts, M. J.,
J. G. Fitz,
A. M. Paradiso,
and
R. C. Boucher.
Multiple modes of regulation of airway epithelial chloride secretion by extracellular ATP.
Am. J. Physiol.
267 (Cell Physiol. 36):
C1442-C1451,
1994
30.
Treharne, K. J.,
L. J. Marshall,
and
A. Mehta.
A novel chloride-dependent, GTP-utilizing protein kinase in plasma membranes from human respiratory epithelium.
Am. J. Physiol.
267 (Lung Cell. Mol. Physiol. 11):
L592-L601,
1994
31.
Treharne, K. J.,
L. J. Marshall,
and
A. Mehta.
Anion-dependent protein kinase activity in human nasal respiratory epithelium in vitro (Abstract).
J. Physiol. (Lond.)
479:
56P,
1994.
32.
Turnheim, K.
Intrinsic regulation of apical sodium entry in epithelia.
Physiol. Rev.
71:
429-445,
1991
33.
Urano, T.,
K. Furukawa,
and
H. Shiku.
Expression of nm23/NDP kinase proteins on the cell surface.
Oncogene
8:
1371-1376,
1993[Medline].
34.
Van Driessche, W.,
and
B. Lindemann.
Concentration dependence of currents through single sodium-selective pores in frog skin.
Nature
282:
519-520,
1979[Medline].
35.
Wagner, P. D.,
and
N. Vu.
Phosphorylation of ATP-citrate lyase by nucleoside diphosphate kinase.
J. Biol. Chem.
270:
21758-21764,
1995
36.
Wang, X.,
Y. Marunaka,
S. Dho,
K. Foskett,
and
H. O'Brodovich.
Activation of chloride currents by intracellular chloride in fibroblasts stably expressing the human cystic fibrosis transmembrane conductance regulator.
Can. J. Physiol. Pharmacol.
71:
645-649,
1993[Medline].
37.
Williams, R. L.,
D. A. Oren,
J. M. Dorado,
S. M. Inouye,
and
E. Arnold.
Crystal structure of Myxococcus xanthus nucleoside diphosphate kinase with a nucleotide substrate at 2.0 Å resolution.
J. Mol. Biol.
235:
1230-1247,
1993.
38.
Xu, L.,
J. Murphy,
and
A. Otero.
Participation of nucleoside-diphosphate kinase in muscarinic K+ channel activation does not involve GTP formation.
J. Biol. Chem.
271:
21120-21125,
1996
This article has been cited by other articles:
|
|
 |
|
  I. Rishal, T. Keren-Raifman, D. Yakubovich, T. Ivanina, C. W. Dessauer, V. Z. Slepak, and N. Dascal Na+ Promotes the Dissociation between Galpha GDP and Gbeta gamma , Activating G Protein-gated K+ Channels J. Biol. Chem., January 31, 2003; 278(6): 3840 - 3845. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Muimo, Z. Hornickova, C. E. Riemen, V. Gerke, H. Matthews, and A. Mehta Histidine Phosphorylation of Annexin I in Airway Epithelia J. Biol. Chem., November 17, 2000; 275(47): 36632 - 36636. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
