Functional Consequences of a Posttransfection Mutation in the H2-H3 Cytoplasmic Loop of the alpha  Subunit of Na,K-ATPase*

(Received for publication, September 19, 1996, and in revised form, December 3, 1996)

Stewart E. Daly Dagger §, Rhoda Blostein Dagger and Lois K. Lane par

From the Dagger  Department of Medicine, McGill University, Montreal, Canada and the par  Department of Pharmacology and Cell Biophysics, University of Cincinnati College of Medicine, Cincinnati, Ohio 45267

ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
Acknowledgments
REFERENCES


ABSTRACT

During kinetic studies of mutant rat Na,K-ATPases, we identified a spontaneous mutation in the first cytoplasmic loop between transmembrane helices 2 and 3 (H2-H3 loop) which results in a functional enzyme with distinct Na,K-ATPase kinetics. The mutant cDNA contained a single G950 to A substitution, which resulted in the replacement of glutamate at 233 with a lysine (E233K). E233K and alpha 1 cDNAs were transfected into HeLa cells and their kinetic behavior was compared. Transport studies carried out under physiological conditions with intact cells indicate that the E233K mutant and alpha 1 have similar apparent affinities for cytoplasmic Na+ and extracellular K+. In contrast, distinct kinetic properties are observed when ATPase activity is assayed under conditions (low ATP concentration) in which the K+ deocclusion pathway of the reaction is rate-limiting. At 1 µM ATP K+ inhibits Na+-ATPase of alpha 1, but activates Na+-ATPase of E233K. This distinctive behavior of E233K is due to its faster rate of formation of dephosphoenzyme (E1) from K+-occluded enzyme (E2(K)), as well as 6-fold higher affinity for ATP at the low affinity ATP binding site. A lower ratio of Vmax to maximal level of phosphoenzyme indicates that E233K has a lower catalytic turnover than alpha 1. These distinct kinetics of E233K suggest a shift in its E1/E2 conformational equilibrium toward E1. Furthermore, the importance of the H2-H3 loop in coupling conformational changes to ATP hydrolysis is underscored by a marked (2 orders of magnitude) reduction in vanadate sensitivity effected by this Glu233 right-arrow Lys mutation.


INTRODUCTION

The Na,K-ATPase is an integral membrane protein complex that catalyzes the exchange of three cytoplasmic sodium ions for two extracellular potassium ions coupled to the hydrolysis of one molecule of ATP. It is a heterodimeric protein comprising a catalytic alpha  subunit and a smaller heavily glycosylated beta  subunit (for reviews, see Refs. 1 and 2). Although an additional protein, gamma , has been found associated with the alpha  and beta  subunits (3), its function is unknown. The Na,K-ATPase is a member of the family of P-type ATPases, which are characteristically phosphorylated by ATP, and the phosphoenzyme intermediate undergoes rapid turnover during the reaction cycle. The phospho- as well as dephosphoenzymes exist in at least two distinct conformational states.

During transport both sodium and potassium ions are occluded within the Na,K-ATPase (for review, see Ref. 4). The nature of the cation occlusion site(s) and the structures involved in gating of these sites to both the cytoplasmic and extracellular milieu are unknown. Approaches using chemical labeling (5, 6) and site-directed mutagenesis (7-12) have identified a number of functionally important carboxyl-containing amino acid residues in transmembrane regions H4, H5, H6, H8, and H9. Point mutations of putative cation binding amino acid residues that resulted in the expression of functional enzymes were characterized by an altered affinity for ATP, K+, and/or Na+.

The prediction that the highly charged amino terminus had a role as a cation gate (13) now seems unlikely. Instead, interaction of this region with other region(s) of the alpha  protein affects the K+ deocclusion pathway of the reaction, probably via alteration in E2/E1 conformational equilibrium. Our recent study (14) demonstrated that the highly charged sequence comprising residues 24-32 of the amino terminus of the alpha  subunit is important in modulating the K+ deocclusion pathway of the reaction cycle. Furthermore, this modulation is not due to the 24-32 nanopeptide per se, but rather to its interaction(s) with other isoform-specific region(s).

This paper describes the characterization of a spontaneous, posttransfection mutation in the H2-H3 cytoplasmic loop of the alpha 1 isoform of the Na,K-ATPase. It differs from the wild type enzyme by having a lysine substituted for glutamic acid in position 233, and is designated E233K. The mutant enzyme is functional and its apparent affinities for sodium and potassium are unaltered. However, it exhibits a decrease in catalytic turnover, an increase in apparent affinity for ATP, and an increase in the rate of conversion of E2(K) to E1.


EXPERIMENTAL PROCEDURES

Recovery and Analysis of Mutant E233K from HeLa Genomic DNA

Two sets of synthetic oligonucleotide primers were prepared for polymerase chain reaction amplification of the 5' and 3' halves of a putative spontaneous mutant in the NH2-terminal chimeric mutant of rat alpha 1 cDNA (alpha 1(1-14alpha 2) cDNA) that had been incorporated into the HeLa genomic DNA in a pRc/CMV (Invitrogen) construct. The 5' primer set included a 32-mer complementary to the sequence of the sense strand of the T7 promoter of pRc/CMV and a 33-mer complementary to the antisense sequence between bases 1828-1860 of the rat alpha 1 cDNA. The 3' primer set included a 28-mer complementary to the sense strand of rat alpha 1 between bases 1684 and 1711 and a 32-mer complementary to an antisense sequence in the SP6 promoter of pRc/CMV at the 3' end of the rat cDNA. The 5' and 3' halves of the alpha 1(1-14alpha 2) cDNA were amplified from 1 µg of HeLa genomic DNA utilizing these primer sets (10 pmol of each primer) and the TaqPlus polymerase reaction kit (Stratagene). Aliquots of the amplified DNA, with or without digestion with HindIII and BamHI, were analyzed on 0.8% agarose gels and compared to products obtained by amplification of 1 ng of purified pRc/CMV-rat alpha 1 plasmid DNA. An aliquot of each reaction mixture was then utilized as template for additional amplification with the same primer sets. Amplified DNA from the 5' and 3' halves of rat alpha 1(1-14alpha 2) was digested with HindIII and BamHI, gel-purified, and ligated into M13mp18/19 that had been digested with HindIII and BamHI. Competent DH5alpha F' Escherichia coli were transformed with each ligation mixture and plated, and the resultant plaques were picked and amplified in liquid cultures. Aliquots of heat-treated supernatants of the M13 cultures were adjusted to 25 mM EDTA, 5% glycerol, and 0.4% SDS and electrophoresed on 0.7% agarose gels to identify M13 clones containing the alpha 1 DNA inserts. Aliquots of the heat-treated supernatants of 37 alpha 5'-containing clones and of 28 alpha 3'-containing clones were pooled and used to infect DH5alpha F' for the preparation of single-stranded M13mp18-5'alpha and M13mp19-3'alpha DNA. The 5' and 3' halves of the rat alpha 1(1-14alpha 2) DNA from the pooled clones were completely sequenced using the Sanger dideoxy method (15), Sequenase version 2.0 (Amersham), and synthetic oligonucleotides as primers.

After detection of a single base substitution at base 950 in the pooled M13mp18-5'alpha DNA, individual clones of M13mp18-5'alpha were sequenced to identify a clone that contained the G950 right-arrow A nucleotide substitution and no other base changes between SalI(875) and BamHI(1780). Double-stranded DNA was then prepared from the identified M13mp18-5'alpha clone, and a SalI(875)-BamHI(1870) fragment containing the G950 right-arrow A substitution was isolated and ligated into both the rat alpha 1 wild type and the rat alpha 1(1-14alpha 2) cDNAs in a modified pIBI30 shuttle vector, in place of the wild type SalI-BamHI fragments. The presence of the mutant sequences and the termini of the exchanged cassettes of these shuttle vector constructs were verified by DNA sequencing. The full-length alpha 1 cDNAs were then isolated by HindIII digestion, gel-purified, and ligated into the HindIII site of pRc/CMV, and their orientations determined by restriction analyses. Twenty µg of QIAGEN-purified expression plasmid DNA was used to transfect HeLa cells with CaPO4, as described by Chen and Okayama (16). HeLa cells expressing the mutant rat alpha 1 proteins were selected by their ability to grow in medium containing 1 µM ouabain, and clones were isolated and amplified as described previously (17).

Membrane Preparations and Enzyme Assays

Membranes were isolated, and assays of Na,K-ATPase, Na-ATPase, and ouabain-sensitive K+(Rb+) influx were carried out as described previously (18). Assays of K+-occluded enzyme and the formation of E1 from K+-occluded enzyme are also described elsewhere (14).


RESULTS

During the course of studies with a rat alpha 1/alpha 2 NH2-terminal chimeric mutant cDNAs transfected into HeLa cells, we observed that the Na,K-ATPase of one clone had functional properties distinct from rat alpha 1. The observation that the enzymic properties of several other clones selected from the same chimeric cDNA (rat alpha 1(1-14alpha 2)) transfection were indistinguishable from those of alpha 1 suggested that a spontaneous, posttranfection mutation had occurred within the coding region of the rat alpha 1 cDNA. As described under "Experimental Procedures," the alteration was identified as the substitution of lysine for glutamate at position 233 in the cytoplasmic region between transmembrane helices H2 and H3. The functional difference between this mutant (E233K) and the rat alpha 1 enzyme was apparent when pump-mediated ATP hydrolysis was measured at micromolar ATP, under which condition ouabain-sensitive Na+-dependent ATPase activity of the alpha 1 isoform is inhibited by K+ (19). Accordingly, although K+ stimulates the dephosphorylation step of the reaction (E2P + K right-arrow E2(K) + Pi) and becomes occluded within the pump protein, its deocclusion is extremely slow. Stimulation of deocclusion is effected by ATP binding with low affinity (E2(K) + ATP right-arrow ATP·E1 + K+; Ref. 19). As a consequence, K+ is an inhibitor of the overall reaction at low ATP, and an activator at high ATP concentration. In fact, as shown previously (14, 18, 20) and described below, at micromolar ATP the response of Na-ATPase to K+ is a sensitive means of characterizing isoform- or mutant-specific differences in the K+ deocclusion pathway of the reaction.

K+ Sensitivity of E233K

As shown in Fig. 1, the Na-ATPase activity of membranes isolated from rat alpha 1-transfected cells is inhibited by the addition of 0.1-5 mM KCl, whereas the activity of E233K is markedly stimulated. At 1 mM KCl, the Na-ATPase activities of the mutant and alpha 1 enzymes are 200% and 50%, respectively, of their control activities measured in the absence of KCl.


Fig. 1. K+ sensitivity of Na-ATPase. ATP hydrolysis was assayed in the presence of 1 µM ATP, 20 mM NaCl, and various concentrations of KCl as described under "Experimental Procedures." Data are presented as percent of Na-ATPase activity (control) measured in the absence of added KCl. Control activities in the presence of 20 mM NaCl, 1 mM MgSO4, 20 mM choline chloride, and 5 µM ouabain to inhibit endogenous activity (nmol/(mg × h)), were 118 ± 41 and 205 ± 48, for alpha 1 and E233K, respectively. open circle , alpha 1; bullet , E233K. Values shown are the means ± S.D. of 11 and 4 experiments for alpha 1 and E233K, respectively.
[View Larger Version of this Image (11K GIF file)]


The change in K+ sensitivity at low ATP concentration caused by the mutation is suggestive of a change in rate of a reaction step following dephosphorylation of the K+-sensitive form of phosphoenzyme commonly referred to as E2P in the Albers-Post model. According to a branched pathway of K+ deocclusion (Scheme I), it is assumed that ATP can bind with either (i) low affinity to the K+-occluded form of the enzyme, E2(K), at a step preceding the release of potassium and the formation of ATP·E1 (pathway a), or (ii) with high affinity at a step following the slow release of potassium from E2(K) (pathway b). Accordingly, potassium inhibition is dependent upon the apparent affinity of the enzyme for ATP at its low affinity site and/or the rate of release of K+ from E2(K). Analysis of kinetic parameters of pathways a and b is described below.


[View Larger Version of this Image (9K GIF file)]

Scheme I.



Kinetic Analysis of the Reaction Modeled According to ATP Binding to Low and High Affinity Sites

The kinetic parameters K'ATP, the apparent affinity for ATP, and Vmax for pathways a and b were obtained from measurements of Na,K-ATPase activity versus ATP concentration varied from 0.1 to 500 µM. Reciprocal plots of the data are shown in Fig. 2, and the kinetic parameters are given in Table I. In the case of the alpha 1 enzyme, the data points can be fitted to a two-component system, each linear in the ranges of 1-8 µM and 25-500 µM ATP. The constants for the apparent affinities for ATP at low and high affinity sites, designated K'L and K'H, respectively, are 331 ± 44 µM and 5.44 ± 1.9 µM for alpha 1; Vmax of the high affinity component, VH, is 0.684 ± 0.20 µmol/(mg × h) and represents 4.2% of the activity of the Vmax of the low affinity component, VL (16.2 ± 3.0 µmol/(mg × h)). These values for alpha 1 are similar to those reported previously (14). In contrast, the reciprocal plot of the E233K mutant is a straight line within the entire range of 1-500 µM, resulting in an apparent affinity of 56.3 ± 14 µM and a Vmax (V'L) of 9.90 ± 1.1 µmol/(mg × h). Although these values are taken from representative experiments and there is some variability in Vmax values, it was observed that the activity of membranes prepared from E233K-transfected cells is generally lower than that of the rat alpha 1-transfected cells.


Fig. 2. Reciprocal plots of Na,K-ATPase activity as a function of ATP concentration. ATP hydrolysis at various concentrations of ATP was determined as described under "Experimental Procedures." Data are presented as 1/(%Vmax). Values of Vmax obtained in the range of 25-500 µM ATP for alpha 1 and 1-500 µM ATP for E233K (µmol/(mg × h)) were 12.8 and 11.2, for alpha 1 and E233K, respectively. Symbols are as described in legend to Fig. 1. Results shown are from a representative experiment; values shown are the average of triplicate determinations.
[View Larger Version of this Image (13K GIF file)]


Table I.

Comparison of kinetic parameters for pathways a and b of ATP hydrolysis catalyzed by alpha 1 and E233K

Values for apparent kinetic constants were determined from the y intercepts and slopes of the low (25-500 µM ATP), high (1-10 µM ATP), and very high (0.1-0.5 µM ATP) affinity components of triphasic reciprocal plots (cf. Fig. 2). Values shown are the means ± S.D. of the number of experiments in parentheses.
 alpha subunit VH/VL K'L K'H K'VH

µM µM µM
 alpha 1 0.042  ± 0.009 (6) 331  ± 44 (6) 5.44  ± 1.9 (6) 0.19  ± 0.09 (3)
E233K VH not detected 56.3  ± 14 (3) Not apparent 0.21  ± 0.01 (2)

An additional component of activity was observed in the range of 0.1-0.5 µM ATP (data not shown), which represented approx 1-2% of VL. The constant for its apparent affinity for ATP, K'VH, was similar for both enzymes (0.19 ± 0.09 µM for alpha 1 and 0.21 ± 0.01 µM for E233K). As suggested previously (14), this very high affinity component may represent an alternate minor pathway of ATP hydrolysis, involving K+ release directly from E2(K), i.e. E2(K) right-arrow E2 + K+ followed by the conversion of E2 to E1. The nature of this minor component was not investigated further.

K+ Occlusion/Deocclusion

The marked difference in effect of K+ on Na-ATPase of E233K and alpha 1 as depicted in Fig. 1 can be attributed to the higher affinity of E233K for ATP (lower K'L) and/or higher rate of K+ deocclusion from E2(K) via pathway b. Since the low specific activity of the Na,K-ATPase expressed in HeLa cell membranes prevents the direct measurement of E2(K) and, therefore, the K+ deocclusion pathway of the reaction, an indirect method was used as described recently (14). In this assay, formation of the K+-occluded enzyme is reflected by the decrease in phosphoenzyme (E32P) formed by phosphorylation with [gamma -32P]ATP following equilibration of the enzyme at room temperature (i) without and (ii) with K+. The reduction in E32P resulting from preincubation with K+ (Delta E32P) is a measure of the amount of E2(K).

As shown in Fig. 3, and in agreement with our previous studies (14), the maximum formation of E2(K) with the alpha 1 isoform is achieved after preincubation of the enzyme with 1 mM KCl, whereas with the alpha 1E233K mutant maximum E2(K) requires at least 4 mM KCl. Using a simple model, Bmax[S]/(K0.5 + [S]), to describe the binding as in earlier studies of K+ occlusion in the kidney enzyme (21), the values of K0.5 obtained are 1.0 ± 0.5 and 0.12 ± 0.03 mM for the E233K and alpha 1 enzymes, respectively (mean ± S.D. of three experiments). Maximum levels of E2(K), expressed as a percentage of phosphoenzyme formed without K+ preincubation (enzyme preincubated without KCl minus the KCl base line) were 93% ± 10% and 97% ± 5% for E233K and alpha 1, respectively.


Fig. 3. Occlusion of K+ at 0 °C. Formation of EP at various concentrations of KCl was determined as described under "Experimental Procedures." Data are presented as percent of maximal E32P, which is the difference of E32P formed in the absence of K+ minus E32P formed in the presence of K+. Values of maximal E32P were (pmol/mg) 20.0 ± 4.8, and 31.5 ± 6.7, for alpha 1 and E233K, respectively. The results were fitted to a simple model of K+ binding (Bmax[S]/(K0.5 + [S])). Symbols are as described in the legend to Fig. 1. Values shown are the means ± S.D. of three experiments.
[View Larger Version of this Image (14K GIF file)]


The shift in the equilibrium E1 + K+ left-right-arrow  E2(K) toward E1 can be accounted for by either a slower rate of occlusion or a faster rate of deocclusion. Although the rate of formation of E1 from E2(K) is not a direct measure of K+ release from E2(K), it is characterized by single exponential kinetics (14) and, as a first approximation, is an estimate of the rate of deocclusion through pathway b.

In the representative experiments shown in Fig. 4, the rate of E1 formation from E2(K) was determined as follows: alpha 1 and E233K enzymes were first equilibrated with 8 mM KCl to form E2(K). Delta E32P, the difference ((E32P formed following preincubation in the absence of K+) minus (E32P formed following preincubation with 8 mM K+)) was taken to represent 100% K+-occluded enzyme (Fig. 3). Deocclusion was measured by following the rate of increase in Delta E32P at 10 °C, a temperature at which deocclusion is sufficiently slow to permit manual assays. We showed previously that (i) maximal E32P formed from E1 was similar at 10 °C and 0 °C, and remained constant over the period used to follow deocclusion at 10 °C, and that Delta E32P remained constant for up to 30 s at 0 °C. Therefore, the time course of increase in Delta E32P at 10 °C reflects the rate of the rate-limiting step in the sequence E2(K) right-arrow E1K right-arrow E1 at that temperature.


Fig. 4. Rate of decrease in K+-occluded enzyme at 10 °C. Formation of E32P was measured at the indicated times following occlusion of K+ as described under "Experimental Procedures." Data are presented as percent of control E32P, which is the difference ((E32P formed in the absence of K+ 0 °C) minus (E32P formed in the presence of 8 mM K+ at 10 °C))/((E32P formed in the absence of K+ 0 °C) minus (E32P formed in the presence of 8 mM K+ at 0 °C)). Symbols are as described in the legend to Fig. 1. Results shown are from a representative experiment; values shown are the average of triplicate determinations.
[View Larger Version of this Image (13K GIF file)]


As can be seen in Fig. 4, the rate of formation of E1P from E2(K) is significantly faster for the E233K mutant compared to the alpha 1 enzyme; rate constants are (s-1) 0.09 ± 0.005 and 0.02 ± 0.004 for E233K and alpha 1, respectively. In other control experiments carried out with alpha 1 and E233K, deocclusion was allowed to proceed for 10 s at 10 °C in ATP-free Na+ medium, after which the E1 formed was measured by rapid dilution and conversion to E1P at 0 °C (5-s phosphorylation at 0 °C). The rate constants estimated from the 10-s decrease in E2(K) were similar to those obtained with 1 µM [gamma -32P]ATP present during deocclusion. This result indicates that the presence of 1 µM ATP during deocclusion did not significantly affect the rate of deocclusion under these conditions.

Phosphoenzyme Turnover and the Effect of Oligomycin

Since oligomycin stabilizes Na+ occlusion in E1 (22, 23) and traps the enzyme in the E1P state, the extent to which EP increases in the presence of oligomycin is a measure of the steady-state distribution of the dephospho- and phosphoenzyme during steady-state catalysis. As shown in Table II (see legend), phosphoenzyme measured at 0 °C in the presence of oligomycin is increased 2.4-fold in the alpha 1 enzyme and only 1.4-fold in E233K. Estimates of turnover, calculated as the ratio of Vmax to EP measured in the presence of oligomycin and Na+, indicate that the E233K mutation results in a 40% reduction in turnover at 37 °C.

Table II.

Comparative turnover and response of phosphoenzyme to oligomycin of alpha 1 and E233K enzymes


 alpha subunit Turnovera Phosphoenzymeb
 - Oligomycin + Oligomycin

min-1 pmol/mg
 alpha 1 7598  ± 1331 (11) 8.69  ± 0.24 19.9  ± 4.1
E233K 4481  ± 196 (3) 19.3  ± 0.68 21.4  ± 1.0

a Values of turnover were obtained from the ratio of Vmax of Na, K-ATPase activity measured at 37 °C in the presence of 100 mM NaCl, 10 mM KCl, and 1 mM ATP, to the maximal amount of EP measured at 0 °C in the presence of 100 mM NaCl, 1 µM ATP, and 20 µg/ml oligomycin.
b Results shown are from a representative experiment; values shown are the average ±S.D. of triplicate determinations. Mean values ±S.D. of the ratio of EP(+oligomycin)/EP(-oligomycin) of at least three independent experiments were 2.4 ± 0.082 and 1.4 ± 0.26, for alpha 1 and E233K, respectively.

Sensitivity to pH

As described by Forbush and Klodos (24), the pH-dependence of Na,K-ATPase is limited partly by the rate of K+ deocclusion at acidic pH, by the rate of the E1P right-arrow E2P transition at neutral pH, and by phosphorylation at pH above pH 8.0. The pH dependence profiles of the E233K mutant and alpha 1 enzymes shown in Fig. 5 indicate that as pH is lowered from the optimum at pH approx  8.0 (room temperature) to pH 6.2, the decrease in activity was less for E233K (20% decrease) than for alpha 1 (50% decrease). In contrast, the activity profile on the alkaline side of the optimum was similar for both enzyme forms.


Fig. 5. pH dependence of Na,K-ATPase activity. Na,K-ATPase activity was measured in the presence of 100 mM NaCl, 10 mM KCl, 3 mM MgSO4, 1 mM ATP, 1 mM EGTA, 30 mM MES (4-morpholineethanesulfonic acid)/Tris (pH 6.5) (or 30 mM Tris/HCl (pH 7.0) or 50 mM Tris/HCl (pH 7.5-8.9)), and 5 µM ouabain (to inhibit endogenous Na,K-ATPase activity) or 10 mM ouabain (base-line activity). The Na,K-ATPase activities corresponding to 100% were (µmol/(mg × h)) 12.2 ± 0.27 and 9.03 ± 0.030 at pH 7.5, for alpha 1 and E233K, respectively. pH values shown were measured at room temperature. Values of activity shown are the means of two representative experiments. Symbols are as described in legend to Fig. 1.
[View Larger Version of this Image (12K GIF file)]


Apparent Cation Affinities

To determine if the substitution of glutamate 233 with a lysine residue alters the apparent affinity of the mutant enzyme for cations, we compared the transport behavior and Na,K-ATPase kinetics of HeLa cells transfected with the E233K and alpha 1 enzymes as described previously (18). Briefly, Rb86 influx sensitive to high (10 mM) ouabain was measured in medium containing either various concentrations of sodium in the presence of monensin to control and maintain intracellular sodium concentration and saturating extracellular KCl concentration, or various concentrations of extracellular KCl, in the presence of monensin and 10 mM NaCl as described previously (25). We found that the apparent affinities for cytoplasmic Na+ (Nacyt) of the E233K mutant and alpha 1 enzymes are indistinguishable. In five separate experiments with the data fitted to a three-site cooperative model, the apparent affinities for Nacyt (K0.5) are (mM) 20.9 ± 2.0 for E233K and 20.1 ± 2.7 for alpha 1. Since intracellular Na+ could not be decreased below approx 15 mM with monensin present (25), we also compared the apparent affinities for Na+ in assays of Na,K-ATPase at 1 mM ATP and constant (10 mM) K+. No difference in apparent affinities for Na+ could be detected. K0.5 values (mM) were 2.49 ± 0.041 for E233K and 2.75 ± 0.16 for alpha 1 (average of three experiments).

As shown earlier by Eisner and Richards (26) in studies of Na,K-ATPase of red cell ghosts, an increase in Kext results in an increase in K'ATP, and increasing the concentration of ATP leads to an increase in K'Kext. Although there is no difference in apparent affinities for either Kext or Nacyt between alpha 1 and E233K pumps under the conditions of the flux assays carried out under physiological conditions of ATP concentration (K0.5 values (mM) were 0.38 ± 0.04 for E233K and 0.37 ± 0.02 for alpha 1 (average of five experiments), it is predicted that as the concentration of ATP is decreased below saturation, the apparent affinity for K+ for both enzymes would increase, and that the increase would be less for E233K than for alpha 1. This prediction held true as shown by the analysis of K+ activation of Na,K-ATPase carried out at 1.0 and 0.1 mM ATP. At 1 mM ATP, with the data fitted to a two-site cooperative model, K0.5 values (mM) for K+ are 1.11 ± 0.034 and 1.76 ± 0.12 for alpha 1 and E233K, respectively; at 0.1 mM ATP, the K0.5 values (mM) are 0.410 ± 0.026 and 1.31 ± 0.074 for alpha 1 and E233K, respectively.

Vanadate Sensitivity

Vanadate interacts with the phosphorylation site in the catalytic H4-H5 domain and inhibits P-type ATPase. Studies with the yeast H+-ATPase have provided evidence that structural alterations in the H2-H3 cytoplasmic loop result in changes in vanadate sensitivity (27, 28). To assess interaction of Glu233 with the catalytic phosphorylation site and/or the role of the region of this mutation on conformational coupling in the Na,K-ATPase, the effect of vanadate was tested. The results shown in Fig. 6 indicate that the Glu233 right-arrow Lys mutation causes a dramatic decrease in sensitivity to vanadate, with Ki values for alpha 1 and E233K being 8 and 2000 µM, respectively.


Fig. 6. Vanadate sensitivity of Na,K-ATPase. ATP hydrolysis at various concentrations of vanadate was determined as described under "Experimental Procedures." Data are presented as percent of Na,K-ATPase activity (control) measured in the absence of vanadate. Control activities in the presence of 100 mM NaCl, 10 mM KCl, 1 mM MgSO4, 5 mM EGTA, 40 mM choline chloride, and 5 µM ouabain (µmol/(mg × h)), were 9.12 ± 0.03 and 8.94 ± 0.19, for alpha 1 and E233K, respectively. The results are from a representative experiment, and the values shown are the means ± S.D. of triplicate determinations. Symbols are as described in the legend to Fig. 1.
[View Larger Version of this Image (14K GIF file)]


Although the insensitivity to vanadate conferred by the Glu233 right-arrow Lys mutation is consistent with the conclusion that this change reflects an alteration in the steady-state E1/E2 distribution as suggested for the yeast H+-ATPase (27, 28), other interpretations are plausible. The effect of this mutation is reminiscent of the decrease in vanadate sensitivity observed in the absence compared to presence of K+ (29), which raises the possibility of a direct alteration in enzyme-K+ interaction. Another possibility is that binding of phosphate is altered. The former is less likely according to the argument that changes in K'K effected by mutations in cytoplasmic domains are probably secondary to changes in K'ATP as discussed below.


DISCUSSION

The experiments described in this report provide evidence for a functional role of the H2-H3 cytoplasmic loop in conformation coupling in the Na,K-ATPase. The changes effected by mutation of glutamate 233 to lysine are consistent with the conclusion that this substitution alters the equilibrium between the major conformational states of the dephospho- and phospho- forms during steady-state catalysis. An increase in the steady-state distribution of E1 and E2 in favor of E1 effected by the Glu233 right-arrow Lys mutation is apparent as (i) an approx 4-fold increase in the rate constant for E1 formation from E2(K), accounting for the higher Km for K+ occlusion of E233K compared to alpha 1, (ii) a 6-fold increase in the apparent affinity for ATP at the step E2(K) + ATP right-arrow ATP·E1 + K+ when the overall Na,K-ATPase reaction is studied at 37 °C, and (iii) a lesser decrease in activity as pH is lowered to pH 6.8 under which condition deocclusion of K+ from E2(K) becomes the main rate-limiting reaction in the wild type enzyme (24). The results are also consistent with a shift in the E1P left-right-arrow  E2P equilibrium toward E1P caused by the Glu233 right-arrow Lys mutation. The evidence is two-fold: the smaller increase in E32P effected by oligomycin and lower turnover of E233K compared to alpha 1, at least at physiological pH under which condition the E1 right-arrow E2P transition is partly rate-limiting (24).

Glu233 is in a region previously identified as having a role in the E1P right-arrow E2P conformational transition (for review, see Ref. 30). When Na,K-ATPase in the E1 conformation is exposed to cleavage at either Leu266 or Arg262 in the H2-H3 cytoplasmic loop, as described by Jorgensen and co-workers (reviewed in Ref. 31), the E1P right-arrow E2P conformational transition is blocked. In the sarcoplasmic reticulum Ca-ATPase, tryptic cleavage at Arg198 as well as site-specific mutation of residues in the predicted beta -strand domain of the H2-H3 loop also blocked the E1P right-arrow E2P conformational transition (reviewed in Ref. 32).

As pointed out by Green and Stokes (33), the H2-H3 loop is well conserved among P-type ATPases. In their model of P-type ion pumps, based largely on studies of the sarcoplasmic reticulum Ca-ATPase (34), their suggestion that the anti-parallel beta -strand is positioned close to the phosphorylation site is supported by vanadate protection of Ca-ATPase from proteolytic degradation (35). In the yeast H+-ATPase, Perlin and co-workers (27, 28) have shown that perturbations of residues in this region not only alters the distribution of conformational intermediates during steady-state catalysis, but also decreases the sensitivity of this P-type pump to vanadate inhibition, consistent also with the conclusion that this region interacts with the catalytic phosphorylation domain.

In studies of Na,K-ATPase, involvement of the H2-H3 loop in structural rearrangements associated with ligand binding and phosphorylation has been apparent in distinctive conformational changes revealed by proteolytic cleavage patterns (36). The importance of these interactions in conformation coupling is emphasized by the remarkable vanadate insensitivity caused by the Glu233 right-arrow Lys mutation. This vanadate insensitivity suggests interaction of Glu233 located in the putative beta -strand region of the H2-H3 loop with the phosphate binding domain within the catalytic H4-H5 loop. It is also possible that the vanadate insensitivity of E233K is the consequence of a decrease in steady-state level of E2 required for Pi (vanadate) binding (cf. the vanadate-insensitive mutants of PMA1; Ref. 28).

When deocclusion of K+ from the K+-occluded state, E2(K), is analyzed as a branched pathway reaction (Scheme I; cf. Refs. 14 and 37), values of Na,K-ATPase activity as a function of varying ATP concentration can be fitted to a biphasic reciprocal plot in the case of the alpha 1 enzyme. The ratio of Vmax values for the high affinity to low affinity components, VH/VL, is 0.043 for alpha 1. In contrast, there is little, if any, evidence of a high affinity component (VH) (pathway b) with the E233K enzyme. This is not unexpected, since, according to the simple Michaelis-Menten relationship shown in Equation 1, activity attributed to pathway a, compared to activity attributed to pathway b, is much greater in the E233K enzyme than it is in the alpha 1 enzyme when ATP is reduced to micromolar concentrations.
v=V<SUB><UP>L</UP></SUB>[<UP>S</UP>]/([<UP>S</UP>]+K′<SUB><UP>L</UP></SUB>)+V<SUB><UP>H</UP></SUB>[<UP>S</UP>]/([<UP>S</UP>]+K′<SUB><UP>H</UP></SUB>) (Eq. 1)
With the E233K mutant, the 6-fold decrease in K'L results in a similar -fold increase in the rate of K+ deocclusion through pathway a, thus largely masking the activity through pathway b. Nevertheless, an increase in the rate of E2(K) right-arrow right-arrow E1 + K+ is also observed with E233K under conditions in which E2(K) is first formed by equilibrating the enzyme with a saturating concentration of K+. The increased rate of K+ deocclusion during subsequent incubation at 10 °C also accounts for the higher K0.5 for K+ under equilibrium binding conditions. It should be noted that the contribution of pathway b to overall activity may increase as temperature is increased. Under the conditions of the deocclusion assays, the relative rate of deocclusion through pathway a compared to that through b at 10 °C is one-half that at 37 °C.1

It is of interest to compare this mutation with those in other cytoplasmic regions of the alpha 1 subunit. In the H4-H5 loop, mutation of aspartate at the phosphorylation site of the Torpedo californica, pig, or sheep kidney enzymes results in complete loss of Na,K-ATPase activity (38-40), and the latter two groups reported an increase in ATP affinity. Mutation of Lys507 of the T. californica enzyme at the putative ATP binding site decreases both activity and ouabain binding capacity (38), and substitution of Cys513 in the rat alpha 1 enzyme decreases ATP affinity (41). In contrast to these mutations, a deletion mutant in which residues 1-32 have been removed from the cytoplasmic amino terminus of alpha 1 results in a fully active enzyme with kinetic changes that are similar to those observed with the E233K mutation (14). This mutation results in a 2.5-fold increase in affinity for ATP at its low affinity binding site and an increase in rate through pathway b (14). With this mutant (alpha 1M32), as with E233K, the rate of conversion of E2(K) to E1 and the apparent affinity for K+ occlusion were increased, and turnover was substantially reduced. Similarly, a NH2-terminal truncated mutant of the Bufo marinus enzyme was recently shown to have a reduced E1 to E2 conformational change (42). As shown elsewhere, the alpha 2 isoform behaves similarly to alpha 1M32 and may be regarded as a conformational variant of alpha 1.

These cytoplasmic mutants contrast with those in which substitutions in transmembrane helices also result in active, functionally altered enzymes, most of which are characterized by changes in affinities for Na+ and K+. Of these, one mutation localized to H5 (Glu779 right-arrow Gln; Ref. 11) alters (decreases) only Na+ affinity. An Asn326 right-arrow Leu mutation in H4 decreased the apparent affinity for Na+ but slightly increased it for K+ (8). Another group of transmembrane substitutions resulted in decreases in affinities for both Na+ and K+. The substituted residues are, in H4: Glu329 right-arrow Gln (43) and Glu327 right-arrow Leu (9); in H5: Glu781 right-arrow Ala (7, 12), with quantitatively smaller changes noted in Glu781 right-arrow Asp (12); and in H6: Thr809 right-arrow Ala (7). In contrast to these mutations, substitution of Ser775 in H5 with either alanine or cysteine decreases apparent K+ affinity dramatically, 31-fold in the case of Ser775 right-arrow Ala and 13-fold in the case of Ser775 right-arrow Cys, with no evidence of a change in Na+ affinity (10).

Substitutions in transmembrane helices effecting an increase in K'K are also associated with an increase in apparent affinity for ATP. In the analysis of Eisner and Richards (26) mentioned earlier, pump-mediated K+ influx was empirically described by a relationship which showed that decreasing Kext concentration (presumably equivalent to increasing K'Kext), increases the apparent affinity for ATP. Similarly, increasing ATP concentration (presumably equivalent to decreasing K'ATP), decreases the apparent affinity for external K+ (and also increases Vmax). Considering, for example, the following sequence of reactions that constitute the low affinity pathway of K+ transport, it is likely that the various transmembrane residues which coordinate K+ ions probably do so with distinct affinities and/or selectivities.

Reaction 1
Accordingly, distinct mutations may differentially alter the rate of specific reaction steps involved in K+ binding, occlusion, and deocclusion, with associated secondary changes in apparent affinity for ATP.

The relationships between cation and ATP affinities of the cytoplasmic mutations are clearly different from those of mutations in transmembrane helices. Although the replacement of glutamate with lysine in the present study and, to a lesser extent, the deletion of the amino terminus described earlier (alpha 1M32) also increased the apparent affinity for ATP, changes in K'K were not observed as long as the ATP concentration was saturating. This behavior is consistent with the interpretation that the primary functional alteration effected by mutating Glu233 to lysine is an increase in the rate of E1 formation from E2(K). We suggest that this substitution in the cytoplasmic H2-H3 loop alters its interaction(s) with other regions of the alpha  subunit resulting in a change in the conformational equilibria, such that the apparent affinity for ATP is increased. Changes in apparent affinity for K+ that were observed at suboptimal ATP concentration are probably secondary to changes in K'ATP, as well as to steps involved in the conversion of conformational forms.


FOOTNOTES

*   This work was supported in part by Grant MT-3876 from the Medical Research Council of Canada (to R. B.), a grant from the Quebec Heart and Stroke Foundation (to R. B.), and National Institutes of Health Grant HL 49204 (to L. K. L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§   Recipient of a postdoctoral fellowship from the Heart and Stroke Foundation of Canada.
   To whom correspondence should be addressed: Montreal General Hospital, 1650 Cedar Ave., Montreal, Quebec H3G 1A4, Canada. Tel.: 514-937-6011 (ext. 4501); Fax: 514-934-8332; E-mail: mirb{at}musica.mcgill.ca.
1   B. Forbush III, unpublished results.

Acknowledgments

We thank Dr. J. B Lingrel, University of Cincinnati College of Medicine, for the alpha 1-transfected HeLa cells, and Dr. B. Forbush III, for communicating unpublished results. We gratefully acknowledge the technical assistance of Rosemarie Scanzano and Apolonia Wilczynski.


REFERENCES

  1. Sweadner, K. J. (1989) Biochim. Biophys. Acta 988, 185-220 [Medline] [Order article via Infotrieve]
  2. Lingrel, J. B, and Kuntzweiler, T. (1994) J. Biol. Chem. 269, 19659-19662 [Free Full Text]
  3. Mercer, R. W., Biemesderfer, D., Bliss, D. P., Jr., Collins, J. H., and Forbush, B., III (1993) J. Cell Biol. 121, 579-586 [Abstract]
  4. Glynn, I. M., and Karlish, S. J. D. (1990) Annu. Rev. Biochem. 59, 171-205 [CrossRef][Medline] [Order article via Infotrieve]
  5. Arguello, J. M., and Kaplan, J. H. (1994) J. Biol. Chem. 269, 6892-6899 [Abstract/Free Full Text]
  6. Goldshleger, R., Tal, D. M., Moorman, J., Stein, W. D., and Karlish, S. J. D. (1992) Proc. Natl. Acad. Sci. U. S. A. 90, 6911-6915
  7. Vilsen, B. (1995) Biochemistry 34, 1455-1463 [Medline] [Order article via Infotrieve]
  8. Vilsen, B. (1995) FEBS Lett. 363, 179-183 [CrossRef][Medline] [Order article via Infotrieve]
  9. Jewell-Motz, E. A., and Lingrel, J. B (1993) Biochemistry 32, 13523-13530 [Medline] [Order article via Infotrieve]
  10. Arguello, J. M., and Lingrel, J. B (1995) J. Biol. Chem. 270, 22764-22771 [Abstract/Free Full Text]
  11. Feng, J., and Lingrel, J. B (1995) Cell. Mol. Biol. Res. 41, 29-37 [Medline] [Order article via Infotrieve]
  12. Koster, J. C., Blanco, G., Mills, P. B., and Mercer, R. W. (1996) J. Biol. Chem. 271, 2413-2421 [Abstract/Free Full Text]
  13. Shull, G. E., Greeb, G., and Lingrel, J. B (1986) Biochemistry 25, 8125-8132 [Medline] [Order article via Infotrieve]
  14. Daly, S. E., Lane, L. K., and Blostein, R. (1996) J. Biol. Chem. 271, 23683-23689 [Abstract/Free Full Text]
  15. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U. S. A. 74, 5463-5467 [Abstract]
  16. Chen, C., and Okayama, H. (1987) Mol. Cell. Biol. 7, 2745-2752 [Medline] [Order article via Infotrieve]
  17. Lane, L. K., Feldmann, J. M., Flarsheim, C. E., and Rybczynski, C. L. (1993) J. Biol. Chem. 268, 17930-17934 [Abstract/Free Full Text]
  18. Daly, S. E., Lane, L. K., and Blostein, R. (1994) J. Biol. Chem. 269, 23944-23948 [Abstract/Free Full Text]
  19. Post, R. L., Hegyvary, C., and Kume, S. (1972) J. Biol. Chem. 247, 6530-6540 [Abstract/Free Full Text]
  20. Wierzbicki, W., and Blostein, R. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 70-74 [Abstract]
  21. Shani, M., Goldshleger, R., and Karlish, S. J. D. (1987) Biochim. Biophys. Acta 904, 13-21 [Medline] [Order article via Infotrieve]
  22. Fahn, S., Koval, G. J., and Albers, R. W. (1966) J. Biol. Chem. 241, 1882-1889 [Abstract/Free Full Text]
  23. Esmann, M., and Skou, J. C. (1985) Biochem. Biophys. Res. Commun. 127, 857-863 [Medline] [Order article via Infotrieve]
  24. Forbush, B., III, and Klodos, I. (1991) in The Sodium Pump: Structure, Mechanism, and Regulation (Kaplan, J. H., and De Weer, P., eds), pp. 211-225, The Rockefeller University Press, New York
  25. Munzer, J. S., Daly, S. E., Jewell-Motz, E. A., Lingrel, J. B, and Blostein, R. (1994) J. Biol. Chem. 269, 16668-16676 [Abstract/Free Full Text]
  26. Eisner, D. A., and Richards, D. E. (1981) J. Physiol. 319, 403-418 [Abstract]
  27. Harris, S. L., Perlin, D. S., Seto-Young, D., and Haber, J. E. (1991) J. Biol. Chem. 266, 24439-24445 [Abstract/Free Full Text]
  28. Bandell, M., Hall, M. J., Wang, G., Seto-Young, D., and Perlin, D. S. (1996) Biochim. Biophys. Acta 1280, 81-90 [Medline] [Order article via Infotrieve]
  29. Sachs, J. R. (1987) J. Gen. Physiol. 90, 291-320 [Abstract]
  30. Andersen, J. P., and Vilsen, B. (1995) FEBS Lett. 359, 101-106 [CrossRef][Medline] [Order article via Infotrieve]
  31. Jorgensen, P. L., and Collins, J. H. (1988) Prog. Clin. Biol. Res. 268A, 85-92
  32. MacLennan, D. H., Clarke, D. M., Loo, T. W., and Skerjanc, S. (1992) Acta Physiol. Scand. 146, 141-150 [Medline] [Order article via Infotrieve]
  33. Green, N. M., and Stokes, D. L. (1992) Acta Physiol. Scand. 146, 59-68
  34. MacLennan, D. H. (1990) Biophys. J. 58, 1355-1365 [Abstract]
  35. Dux, L., and Martonosi, A. (1983) J. Biol. Chem. 258, 10111-10115 [Abstract/Free Full Text]
  36. Lutsenko, S., and Kaplan, J. H. (1994) J. Biol. Chem. 269, 4555-4564 [Abstract/Free Full Text]
  37. Sachs, J. R. (1994) Biochim. Biophys. Acta 1193, 199-211 [Medline] [Order article via Infotrieve]
  38. Ohtsubo, M., Noguchi, S., Takeda, K., Morohashi, M., and Kawamura, M. (1990) Biochim. Biophys. Acta 1021, 157-160 [Medline] [Order article via Infotrieve]
  39. Kuntzweiler, T. A., Wallick, E. T., Johnson, C. L., and Lingrel, J. B (1995) J. Biol. Chem. 270, 16206-16212 [Abstract/Free Full Text]
  40. Pedersen, P. A., Rasmussen, J. H., and Jorgensen, P. L. (1996) J. Biol. Chem. 271, 2514-2522 [Abstract/Free Full Text]
  41. Lane, L. K. (1993) Biochem. Mol. Biol. Int. 31, 817-832 [Medline] [Order article via Infotrieve]
  42. Wang, X., Jaisser, F., and Horisberger, J. D. (1996) J. Physiol. 491, 579-594 [Abstract]
  43. Vilsen, B. (1993) Biochemistry 32, 13340-13349 [Medline] [Order article via Infotrieve]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.