HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 278, Issue 22, 20245-20258, May 30, 2003

This Article Abstract Full Text (PDF) All Versions of this Article: 278/22/20245    most recent M301122200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Clausen, J. D. Articles by Andersen, J. P. Search for Related Content PubMed PubMed Citation Articles by Clausen, J. D. Articles by Andersen, J. P. Social Bookmarking                      What's this?

Importance of Conserved N-domain Residues Thr⁴⁴¹, Glu⁴⁴², Lys⁵¹⁵, Arg⁵⁶⁰, and Leu⁵⁶² of Sarcoplasmic Reticulum Ca²⁺-ATPase for MgATP Binding and Subsequent Catalytic Steps

PLASTICITY OF THE NUCLEOTIDE-BINDING SITE^*

Johannes D. Clausen , David B. McIntosh  ¶, Bente Vilsen , David G. Woolley  and Jens Peter Andersen  ||

From the Department of Physiology, University of Aarhus, DK-8000 Aarhus C, Denmark and the Division of Chemical Pathology, Department of Clinical Laboratory Sciences, Faculty of Health Sciences, University of Cape Town, Cape Town 7925, South Africa

Received for publication, February 3, 2003

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Nine single mutations were introduced to amino acid residues Thr⁴⁴¹, Glu⁴⁴², Lys⁵¹⁵, Arg⁵⁶⁰, Cys⁵⁶¹, and Leu⁵⁶² located in the nucleotide-binding domain of sarcoplasmic reticulum Ca²⁺-ATPase, and the functional consequences were studied in a direct nucleotide binding assay, as well as by steady-state and transient kinetic measurements of the overall and partial reactions of the transport cycle. Some partial reaction steps were also examined in mutants with alterations to Phe⁴⁸⁷, Arg⁴⁸⁹, and Lys⁴⁹². The results implicate all these residues, except Cys⁵⁶¹, in high affinity nucleotide binding at the substrate site. Mutations Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Leu⁵⁶² Phe were more detrimental to MgATP binding than to ATP binding, thus pointing to a role for these residues in the binding of Mg²⁺ or to a difference between the interactions with MgATP and ATP. Subsequent catalytic steps were also selectively affected by the mutations, showing the involvement of the nucleotide-binding domain in these reactions. Mutation of Arg⁵⁶⁰ inhibited phosphoryl transfer but enhanced the E₁PCa₂ E₂P conformational transition, whereas mutations Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, Lys⁴⁹² Leu, and Lys⁵¹⁵ Ala inhibited the E₁PCa₂ E₂P transition. Hydrolysis of the E₂P phosphoenzyme intermediate was enhanced in Glu⁴⁴² Ala, Lys⁴⁹² Leu, Lys⁵¹⁵ Ala, and Arg⁵⁶⁰ Glu. None of the mutations affected the low affinity activation by nucleotide of the phosphoenzyme-processing steps, indicating that modulatory nucleotide interacts differently from substrate nucleotide. Mutation Glu⁴⁴² Ala greatly enhanced reaction of Lys⁵¹⁵ with fluorescein isothiocyanate, indicating that the two residues form a salt link in the native protein.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
The sarcoplasmic reticulum Ca²⁺-ATPase¹ is an energy-transducing enzyme responsible for active transport of Ca²⁺ from the cytosol into the lumen of sarcoplasmic reticulum. It is an integral 110-kDa membrane protein belonging to the family of P-type ATPases, in which autophosphorylation by ATP of a conserved aspartic acid residue is coupled to conformational changes resulting in ion translocation (Scheme 1). The high resolution x-ray structure of the Ca²⁺-ATPase crystallized in a Ca²⁺-bound dephospho form (presumably corresponding to E₁Ca₂ in Scheme 1) has revealed that the enzyme is made up of 10 membrane-spanning helices and a large cytoplasmic head that binds ATP and catalyzes its hydrolysis to yield ADP and P_i. The head consists of three well defined domains: the nucleotide-binding domain ("domain N," supposed to bind the adenosine part of ATP), phosphorylation domain ("domain P" with Asp³⁵¹, which is phosphorylated), and actuator domain ("domain A," believed to undergo large movements during the reaction cycle) (1). The catalytic site may, at different stages of the transport cycle, consist of residues contributed from all three head domains, and many questions regarding the nature of the nucleotide-binding site(s) are unresolved.

View larger version (13K): [in this window] [in a new window]   SCHEME 1. Ca2+-ATPase reaction cycle. Major conformational changes and ligand binding and dissociation steps are shown. Boxed ATP indicates steps for which the rate is enhanced by additional binding of ATP or MgATP that is not hydrolyzed ("modulatory ATP").

In domain N, labeling of Lys⁵¹⁵ with FITC (2, 3) or Lys⁴⁹² with TNP-8N₃-ATP (4) is competitive with respect to ATP binding at the catalytic site. The specific photolabeling of Lys⁴⁹² with TNP-8N₃-ATP has permitted measurement of the binding of nucleotide to mutant Ca²⁺-ATPases in a competition assay and identified Phe⁴⁸⁷, Arg⁴⁸⁹, and Lys⁴⁹² as likely nucleotide-binding residues (5). The location of TNP-AMP in the crystal structure, obtained from a difference Fourier map following soaking of the E₁Ca₂ crystals in a solution containing TNP-AMP, supported involvement of these residues in the binding of nucleotide, and in addition revealed the proximity of TNP-AMP to several other residues, including Thr⁴⁴¹, Glu⁴⁴², Lys⁵¹⁵, Arg⁵⁶⁰, and Leu⁵⁶² (1).

A surprising feature of the crystal structure of the Ca²⁺-bound E₁Ca₂ dephospho form is that the configuration of the cytoplasmic domains is quite open with the nucleotide-binding site being separated from Asp³⁵¹ in domain P by more than 25 Å, implying large domain movements during ATP binding and hydrolysis (1). Presumably, the binding of nucleotide leads to closing of the structure. The movements contribute to uncertainties about the nucleotide-binding residues and the orientation of the bound nucleotide. Although the high affinity binding sites for TNP-nucleotide and ATP seem to overlap (6), they are not necessarily identical. Thapsigargin, a potent inhibitor of the Ca²⁺-ATPase, has no effect on TNP-ATP or TNP-8N₃-ATP binding, and yet decreases the affinity of the pump for ATP by at least 100-fold (5, 7). The doubts with respect to the nucleotide site are exacerbated by functional studies showing that ATP in addition to being the phosphorylating substrate exerts modulatory effects on various steps of the pump cycle, see boxed ATP in Scheme 1 (8, 9, 10, 11, 12, 13, 14, 15, 16). In view of the predicted large domain movements and presence of the phosphoryl group at the catalytic site during part of the cycle, it seems unlikely that modulatory ATP binds exactly as substrate ATP does.

In the present study, we have introduced mutations of Thr⁴⁴¹, Glu⁴⁴², Lys⁵¹⁵, Arg⁵⁶⁰, Cys⁵⁶¹, and Leu⁵⁶², and we have also extended the previous study (5) on mutants with alterations to Phe⁴⁸⁷, Arg⁴⁸⁹, and Lys⁴⁹². With the exception of Cys⁵⁶¹, these residues show a high degree of conservation throughout the P-type ATPase family (17). A recent study, applying solution nuclear magnetic resonance techniques to a 28-kDa recombinant Ca²⁺-ATPase fragment corresponding to domain N pinpointed Thr⁴⁴¹ and Glu⁴⁴² as the two residues providing the largest shifts in the backbone¹⁵N spectra upon addition of a non-hydrolyzable ATP analogue to the medium (18). Marked spectral shifts were also seen for residues close in the linear sequence to Lys⁴⁹², Lys⁵¹⁵, and Arg⁵⁶⁰. Site-directed mutagenesis studies have previously shown the importance of Lys⁵¹⁵ and Arg⁵⁶⁰ in the overall ATPase function (19, 20, 21). Mutation of Lys⁵¹⁵ to arginine, alanine, glutamine, and glutamate resulted in 65, 30, 25, and 5% Ca²⁺-transport activity relative to wild type (19, 20). Mutation of Arg⁵⁶⁰ to alanine abolished phosphorylation from ATP (21). Thr⁴⁴¹, Glu⁴⁴², Cys⁵⁶¹, and Leu⁵⁶², or their homologues in other P-type ATPases, have not been subjected previously to mutational analysis.

This study, presenting the first direct measurements of nucleotide-binding properties of mutants with alterations to Thr⁴⁴¹, Glu⁴⁴², Lys⁵¹⁵, Arg⁵⁶⁰, Cys⁵⁶¹, and Leu⁵⁶² of the Ca²⁺-ATPase, reveals marked effects on nucleotide binding to the substrate site of the dephosphoenzyme, except in the case of the Cys⁵⁶¹ mutants. Moreover, our analysis of other steps in the pump cycle (Scheme 1) demonstrates that phosphorylation, the Ca²⁺-translocating E₁PCa₂ E₂P conformational transition, as well as the hydrolysis of the E₂P phosphoenzyme intermediate are affected to various extents by the mutations. The low affinity modulatory effects of ATP on the phosphoenzyme-processing steps are, however, not significantly affected. Finally, we provide functional evidence for close interaction between Lys⁵¹⁵ and Glu⁴⁴² in experiments demonstrating direct involvement of Glu⁴⁴² in the labeling reaction of Lys⁵¹⁵ with FITC.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Mutagenesis, Expression, and Assays of Overall Function—Oligonucleotide-directed mutagenesis of cDNA encoding the rabbit fast twitch muscle Ca²⁺-ATPase (SERCA1a isoform) was carried out as described previously (22). For expression, the wild-type or mutant cDNA, inserted in the pMT2 vector (23), was transfected into COS-1 cells using the calcium phosphate precipitation method (24). The microsomal fraction containing expressed wild-type or mutant Ca²⁺-ATPase was isolated by differential centrifugation (19). The concentration of expressed Ca²⁺-ATPase was quantified by a specific enzyme-linked immunosorbent assay (25) or by determination of the maximum capacity for phosphorylation by inorganic phosphate in the presence of 30% (v/v) dimethyl sulfoxide ("active site concentration," see Ref. 26). Transport of⁴⁵Ca²⁺ into the microsomal vesicles was measured by filtration, and the ATPase activity was measured by determining the amount of P_i liberated as described previously (26).

Phosphorylation from [-³²P]ATP and³²P_i—Manual mixing experiments at various buffer and temperature conditions (detailed in the figure legends) were carried out according to the principles described previously (22, 25, 26). Transient kinetic experiments at 25 °C were performed using a Bio-Logic quench-flow module QFM-5 (Bio-Logic Science Instruments, Claix, France) as described (27). In all phosphorylation experiments, acid quenching was performed with 0.5–2 volumes of 25% (w/v) trichloroacetic acid containing 100 mM H₃PO₄. The acid-precipitated protein was washed by centrifugation and subjected to SDS-PAGE in a 7% polyacrylamide gel at pH 6.0 (28), and the radioactivity associated with the separated Ca²⁺-ATPase band was quantified by imaging, using a Packard Cyclone^TM Storage Phosphor System. Background phosphorylation levels, subtracted from all data points, were usually determined in parallel experiments with control microsomes isolated from mock-transfected COS-1 cells. In some of the dephosphorylation experiments, the constant phosphorylation level reached after the exponential decay was taken as background (usually 5% of the initial phosphorylation).

To remove contaminant ADP and AMP, the non-radioactive ATP added in millimolar concentration in dephosphorylation experiments was purified by ion exchange chromatography on a Sephadex DEAE-A25 column before use. By using an NADH-coupled spectrophotometric assay with phosphoenolpyruvate, lactate dehydrogenase, and pyruvate kinase, the purified ATP preparation was found to contain less ADP than the detection limit of 0.1 mol %.

Assays for Nucleotide Binding—The synthesis of [-³²P]TNP-8N₃-ATP, the photolabeling of COS-1 cell microsomes containing wild-type or mutant Ca²⁺-ATPase, the inhibition by ATP, and the quantification of labeled bands by electronic autoradiography following SDS-PAGE were carried out as described previously (5, 29). Generally, the concentration of [-³²P]TNP-8N₃-ATP was 3x K_0.5 in the inhibition experiments with ATP (where K_0.5 is ligand concentration giving half-maximum effect). Details of the buffer composition are given in the legends. CrATP-induced Ca²⁺ occlusion was measured as described previously (5, 30).

Calculations and Data Analysis—The ion concentrations in the reaction buffers were calculated using the program WEBMAXC, available on the World Wide Web, and the stability constants therein (31). Generally, experiments were conducted at least twice, and the complete set of data was analyzed by nonlinear regression using the SigmaPlot program (SPSS, Inc.). Monoexponential functions were fitted to the phosphorylation and dephosphorylation time courses. The analysis of ligand concentration dependences was based on the Hill equation,

(Eq. 1)

For analysis of the [-³²P]TNP-8N₃-ATP labeling data, a constant or linear component was added to represent nonspecific labeling as described (5), and the Hill coefficient was set to 1. The "true" dissociation constant for ATP and MgATP binding was calculated using the previously validated equation for competitive inhibition of the [-³²P]TNP-8N₃-ATP labeling (5).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Mutagenesis and Expression—Nine Ca²⁺-ATPase constructs with mutations of potential nucleotide-binding residues were prepared in this study. Thr⁴⁴¹, Glu⁴⁴², and Lys⁵¹⁵ were replaced individually with alanine. Cys⁵⁶¹ was replaced with alanine or tryptophan; Arg⁵⁶⁰ was replaced with leucine, valine, or glutamate; and Leu⁵⁶² was replaced with phenylalanine. The latter substitution was chosen to test the importance of the leucine for nucleotide binding by either introducing steric hindrance at this position or possibly enhancing the affinity for nucleotide by creating an aromatic "sandwich" for the adenine ring with Phe⁴⁸⁷. All mutants could be expressed in COS-1 cells at levels similar to that of the wild type (data not shown), allowing studies of the overall function as well as the partial reactions of the enzyme cycle. In addition, some supplementary studies of partial reactions were carried out with previously constructed mutants (5) with alterations to Phe⁴⁸⁷, Arg⁴⁸⁹, and Lys⁴⁹².

Overview of Functional Effects—Fig. 1A shows ATP-driven ⁴⁵Ca²⁺ accumulation in microsomal vesicles measured in the presence of oxalate to trap Ca²⁺ in the lumen. The MgATP concentration was 5 mM, i.e. several hundredfold higher than required to saturate the phosphorylation reaction in wild type. All the mutants were able to transport Ca²⁺, albeit at reduced rates compared with wild type. The most significant reduction was seen for mutant Arg⁵⁶⁰ Glu with a Ca²⁺ transport rate corresponding to 17% that of the wild type, whereas 80% transport was obtained for the Cys⁵⁶¹ mutants. The Ca²⁺ transport rates of the remaining six mutants were roughly half (39–66%) that of wild type.

View larger version (21K): [in this window] [in a new window]   FIG. 1. Initial overview of ATP-driven Ca2+ transport (A), ATPase activity (B), and phosphorylation from ATP (C). A, measurements of ATP-driven Ca2+ transport were performed by filtration following incubation for 5 min at 37 °C in a medium containing 20 mM MOPS/Tris (pH 6.8), 100 mM KCl, 5 mM MgATP, 5 mM potassium oxalate (included to trap Ca2+ inside the microsomal vesicles), 0.5 mM EGTA, and 0.45 mM 45CaCl2. The Ca2+ transport activity of the wild type was taken as 100%, and each mutant was related to this level following correction for the variation in expression level. B, the rate of Ca2+-activated ATP hydrolysis was determined at 37 °C in the presence (gray bars) or absence (black bars) of 1 µM Ca2+ ionophore A23187 [GenBank] in a medium containing 50 mM TES/Tris (pH 7.0), 100 mM KCl, 7 mM MgCl2, 1 mM EGTA, 0.9 mM CaCl2 (giving a free Ca2+ concentration of 3 µM), and 5 mM ATP. Following subtraction of the background activity determined in the absence of Ca2+, the molecular activity (catalytic turnover rate) shown was calculated as the amount of Pi liberated per Ca2+-ATPase molecule/s. C, the phosphorylation with [-32P]ATP was carried out for 5 s at 25 °C in a medium containing 40 mM MOPS/Tris (pH 7.0), 80 mM KCl, 5 mM MgCl2, 100 µM CaCl2, and either 5 µM [-32P]ATP (gray bars) or 50 µM [-32P]ATP (black bars). The phosphorylation level of the wild type incubated in a medium containing 100 mM MES/Tris (pH 6.0), 10 mM MgCl2, 2 mM EGTA, 30% (v/v) dimethyl sulfoxide, and 0.5 mM 32Pi for 10 min at 25 °C was taken as 100% ("active site concentration"), and all other values were related to this level following correction for the variation in expression level. Standard errors are indicated by the error bars on the columns.

Fig. 1B shows the ATPase activity in the presence of 5 mM MgATP without oxalate, in the absence and presence of the calcium ionophore A23187 [GenBank] . In the absence of ionophore (conditions resulting in net Ca²⁺ uptake in the vesicles), a similar pattern was obtained for ATP hydrolysis as for Ca²⁺ transport (Fig. 1B, black columns), Arg⁵⁶⁰ Glu showing the most marked reduction of turnover rate, the Cys⁵⁶¹ mutants being wild type like, and the remaining six mutants hydrolyzing ATP at turnover rates of 40–58% as compared with wild type. The good correlation between the turnover rate for ATP hydrolysis and ATP-driven accumulation of Ca²⁺ in the microsomes demonstrates that the coupling between ATP hydrolysis and Ca²⁺ transport was retained in the mutants. Addition of calcium ionophore A23187 [GenBank] to the reaction medium increased the rate of ATP hydrolysis in the wild type 2–3-fold because of relief of the "back inhibition" of the rate-limiting E₁PCa₂ E₂P transition imposed by Ca²⁺ accumulated at high concentrations in the microsomal vesicles. A similar increase was seen for Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, Lys⁵¹⁵ Ala, and the Cys⁵⁶¹ mutants, whereas for Leu⁵⁶² Phe and the three Arg⁵⁶⁰ mutants the ATPase activity remained essentially unaltered (Fig. 1B, gray columns). The latter result implies that a reaction step other than the E₁PCa₂ E₂P transition is rate-limiting in the case of Leu⁵⁶² Phe and Arg⁵⁶⁰ mutants.

In the Ca²⁺-bound E₁Ca₂ state, the wild-type Ca²⁺-ATPase forms a phosphoenzyme intermediate by reaction with MgATP (cf. Scheme 1). Fig. 1C shows the results of experiments where phosphorylation of Ca²⁺-saturated enzyme was carried out for 5 s at 25 °C with either 5 or 50 µM [-³²P]MgATP. For wild type, the maximum steady-state phosphoenzyme level of 80% of the total active-site concentration is close to being reached at 5 µM MgATP; also for the Cys⁵⁶¹ mutants, there was not much difference between 5 and 50 µM MgATP. By contrast, mutants Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, Lys⁵¹⁵ Ala, and in particular Arg⁵⁶⁰ Leu, Arg⁵⁶⁰ Val, and Leu⁵⁶² Phe showed much lower phosphoenzyme levels with 5 µM MgATP than with 50 µM MgATP, indicating a markedly reduced apparent affinity for the nucleotide, relative to wild type. Mutant Arg⁵⁶⁰ Glu was unable to phosphorylate significantly above the background level at either concentration of MgATP (Fig. 1C).

MgATP Dependence and Time Course of Phosphorylation from [-³²P]ATP—The MgATP concentration dependence of phosphorylation was further investigated at 0 °C, where the ATP hydrolysis rate was sufficiently low to allow the phosphorylation to be studied at submicromolar concentrations of the substrate. As seen in Fig. 2A, a 10–120-fold reduction of apparent affinity for MgATP (increase of K_0.5) relative to wild type was found for mutants Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, Lys⁵¹⁵ Ala, and Leu⁵⁶² Phe; and for mutants Arg⁵⁶⁰ Leu and Arg⁵⁶⁰ Val, the K_0.5 for MgATP was at least 3 orders of magnitude lower than that of wild type.

View larger version (23K): [in this window] [in a new window]   FIG. 2. MgATP concentration dependence (A) and time course (B) of phosphorylation from ATP. A, wild type and mutants were phosphorylated at 0 °C for 15 s in a medium containing 40 mM MOPS/Tris (pH 7.0), 80 mM KCl, 5 mM MgCl2, 100 µM CaCl2, and varying concentrations of [-32P]ATP as indicated. The lines show the best fits of the Hill equation with the Hill coefficient set to 1, giving the K0.5 values (in µM ± S.E.) indicated in parentheses: open circles, wild type, (0.067 ± 0.005); open squares, Thr441 Ala (6.2 ± 0.6); open triangles, Glu442 Ala (0.72 ± 0.06); reversed solid triangles, Lys515 Ala (1.6 ± 0.2); open diamonds, Arg560 Leu (>>100); solid triangles, Arg560 Val (100); and solid circles, Leu562 Phe (8.1 ± 1.4). The 100% value corresponds to the phosphorylation level reached at infinite ATP concentration as deduced from the fit. B, phosphorylation was carried out at 25 °C in a medium containing 40 mM MOPS/Tris (pH 7.0), 80 mM KCl, 100 µM CaCl2, 5 mM MgCl2, and 5 µM [-32P]ATP, and the samples were acid-quenched at serial time intervals, using a Bio-Logic QFM-5 quench-flow module with mixing protocol as described previously (34). In each case, the level of phosphorylation after 5 s was taken as 100% (for a comparison of the specific phosphorylation levels of wild type and mutants after 5 s, see Fig. 1C). The lines show the best fits of a monoexponential function, giving the rate constants (in s-1 ± S.E.) indicated in parentheses: open circles, wild type (59.7 ± 2.0); open squares, Thr441 Ala (5.3 ± 0.6); open triangles, Glu442 Ala (13.3 ± 0.6); reversed solid triangles, Lys515 Ala (11.1 ± 0.8).

Fig. 2B shows the time course of phosphorylation at 25 °C of wild type and selected mutants determined by quench-flow methodology in the presence of 5 µM [-³²P]MgATP under conditions identical to those corresponding to Fig. 1C. Leu⁵⁶² Phe and the three Arg⁵⁶⁰ mutants were not included in these studies because their phosphorylation levels were too low to allow reliable pre-steady-state measurements (cf. Fig. 1C). The wild type reaches a steady-state level of phosphorylation within a few hundred milliseconds, with a slight initial over-shoot (Fig. 2B, cf. also Ref. 27). For simplicity, a monoexponential function was fitted to the data, giving an apparent rate constant k_obs = 60 s^-1 for the approach to steady state. In comparison, 5–12-fold lower k_obs values were determined for mutants Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Lys⁵¹⁵ Ala (Fig. 2B). The k_obs value decreased in the same order as the apparent MgATP affinity observed in Fig. 2A (Glu⁴⁴² Ala > Lys⁵¹⁵ Ala > Thr⁴⁴¹ Ala).

MgATP Dependence of ATP Hydrolysis—For wild-type Ca²⁺-ATPase, the MgATP activation profile of ATPase activity has a complex appearance, because MgATP (or ATP) in addition to being the phosphorylating substrate exerts modulatory effects on various steps of the pump cycle, cf. Scheme 1 (8, 9, 10, 11, 12, 13, 14, 15, 16). The basal activation to about 20% of the maximum activity occurring at MgATP concentrations below 10 µM in wild type reflects the binding of MgATP to the E₁Ca₂ form as phosphorylating substrate. It can be seen in Fig. 3 that for mutant Leu⁵⁶² Phe and the three Arg⁵⁶⁰ mutants, the basal activation was shifted to much higher MgATP concentrations in agreement with the data in Fig. 2A. Less drastic right-shifts of the basal activation were observed for mutants Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Lys⁵¹⁵ Ala with plateau levels being reached at about 50 µ_M MgATP, whereas the Cys⁵⁶¹ mutants were wild type-like. At least two secondary activation phases occur for the wild-type enzyme, between 10 and 200 µM MgATP and at higher MgATP concentrations, reflecting the accelerating effect of nucleotide on partial reaction steps preceding and subsequent to phosphorylation (9, 10, 12, 13, 14, 15, 16). Fig. 3 shows that the Cys⁵⁶¹ mutants displayed a wild type-like secondary activation. For mutant Leu⁵⁶² Phe and the three Arg⁵⁶⁰ mutants, the severely reduced ATPase activity prevented a clear distinction between the basal and the intermediate modulatory phases (Fig. 3), but a distinct low affinity activation phase similar to that of the wild type is seen above 200 µM MgATP for Arg⁵⁶⁰ Leu, Arg⁵⁶⁰ Val, and Leu⁵⁶² Phe, as well as for Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Lys⁵¹⁵ Ala, indicating that nucleotide binding with low affinity accelerates at least one partial reaction in these mutants, although in contrast to the wild type the activation profiles level off above 3 mM MgATP for Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Lys⁵¹⁵ Ala. Furthermore, taking into consideration that the basal activation is shifted to higher MgATP concentrations relative to wild type, the activation occurring with intermediate affinity, between 10 and 200 µM MgATP in wild type, seems to be less pronounced or right-shifted in Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Lys⁵¹⁵ Ala. This is most obvious for Glu⁴⁴² Ala, where the ATPase activity was nearly constant between 50 and 200 µM MgATP (Fig. 3).

View larger version (22K): [in this window] [in a new window]   FIG. 3. MgATP dependence of ATPase activity. The rate of ATP hydrolysis was measured at 37 °C in a medium containing 50 mM TES/Tris (pH 7.5), 100 mM KCl, 100 µM CaCl2, 1 µM Ca2+ ionophore A23187 [GenBank] , and varying concentrations of Mg2+ and ATP to give the indicated MgATP concentrations and 1 mM free Mg2+. Following subtraction of the background activity determined in the absence of Ca2+ (presence of 1 mM EGTA without CaCl2), the catalytic turnover rate shown was calculated as the amount of Pi liberated per Ca2+-ATPase molecule/s. The symbols are as follows: open circles, wild type; reversed open triangles, Cys561 Ala; solid squares, Cys561 Trp; solid triangles, Arg560 Val; open diamonds, Arg560 Leu; solid diamonds, Arg560 Glu; open squares, Thr441 Ala; open triangles, Glu442 Ala; reversed solid triangles, Lys515 Ala and solid circles, Leu562 Phe.

Acetyl Phosphate-driven Ca²⁺ Uptake—As an alternative to ATP, acetyl phosphate can be used as an energy source to drive Ca²⁺ transport into the microsomal vesicles (16). Fig. 4 shows the time course for wild type and the mutants. For most mutants, the decrease of the transport rate relative to wild type was rather similar to the decrease seen for ATP-driven transport at 5 mM MgATP (Fig. 1A), as also shown previously for mutant Lys⁵¹⁵ Ala (20), indicating that at high MgATP concentration (Fig. 1, A and B) the limitation of the overall reaction rate seen for the mutants is not imposed by binding defects specific to the nucleotide, but rather by defects in other partial reaction steps. Interestingly, mutant Leu⁵⁶² Phe is a clear exception to this pattern. This mutant showed a markedly reduced ATP-driven Ca²⁺ transport (Fig. 1A) and ATPase activity (Fig. 1B), but wild type-like acetyl phosphate-dependent Ca²⁺ transport (Fig. 4).

View larger version (21K): [in this window] [in a new window]   FIG. 4. Acetyl phosphate-driven Ca2+ transport. Transport activity of COS-1 cell microsomes was measured at 25 °C in 25 mM MOPS/Tris (pH 6.8), 100 mM KCl, 5 mM MgCl2, 0.55 mM 45CaCl2, 0.5 mM EGTA, 5 mM potassium oxalate, and 10 mM acetyl phosphate for the times shown. The ordinate shows the amount of Ca2+ accumulated per mg of Ca2+-ATPase protein. The symbols are as follows: open circles, wild type; open squares Thr441 Ala; open triangles, Glu442 Ala; reversed solid triangles, Lys515 Ala; solid triangles, Arg560 Val; open diamonds, Arg560 Leu; solid diamonds, Arg560 Glu; reversed open triangles, Cys561 Ala; and solid circles, Leu562 Phe.

Ca²⁺ Dependence—To examine whether a reduced affinity for Ca²⁺ could be involved in the observed effects of the mutations on Ca²⁺ transport and ATPase activity, we performed a Ca²⁺ titration of the ATPase activity. The Ca²⁺ concentration giving half-maximum activation of ATP hydrolysis was 0.33 ± 0.01 µM for the wild type and 0.64 ± 0.05, 0.41 ± 0.02, 0.35 ± 0.02, 0.33 ± 0.02, 0.54 ± 0.04, 0.38 ± 0.02, 0.47 ± 0.04, and 0.38 ± 0.03 µM for Arg⁵⁶⁰ Leu, Arg⁵⁶⁰ Val, Cys⁵⁶¹ Ala, Cys⁵⁶¹ Trp, Leu⁵⁶² Phe, Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Lys⁵¹⁵ Ala, respectively (the K_0.5 value was obtained by fitting the Hill equation, ± S.E., from the regression analysis, data not shown). Thus, although significant alterations to Ca²⁺ binding are evident for mutants Arg⁵⁶⁰ Leu and Leu⁵⁶² Phe, the mutants were >80% saturated with Ca²⁺ at the free Ca²⁺ concentration of 3 µ_M present during the experiments corresponding to Fig. 1, A and B, and close to 100% saturated at the higher free Ca²⁺ concentrations present during the experiments corresponding to Fig. 1C, Fig. 2, A and B, Fig. 3, and Fig. 4. Hence, the observed effects of the mutations cannot be ascribed to variable Ca²⁺ saturation.

Nucleotide Binding—To examine the mutational effects on nucleotide binding directly, we applied a photolabeling assay that takes advantage of the highly specific labeling of Lys⁴⁹² with [-³²P]TNP-8N₃-ATP (5). In this assay, the affinities for the free and Mg²⁺-bound forms of [-³²P]TNP-8N₃-ATP are determined from the dependence of photolabeling on the concentration of the label in the absence and presence of Mg²⁺, respectively. By studying competitive inhibition of photolabeling in the presence of varying concentrations of ATP and MgATP (the latter often being referred to as the "true" substrate (32)), it is also possible to obtain highly accurate values for the K_D corresponding to these nucleotides (5, 33, 34). Photolabeling is carried out in the absence of Ca²⁺ to avoid phosphorylation of the Ca²⁺-ATPase and hydrolysis of the photolabel, and at pH 8.5 to reduce nonspecific labeling and ensure that the predominant enzyme conformation is E₁ (5). Results of photolabeling experiments with the nine mutants in the absence and presence of Mg²⁺ are summarized in Table I. Also presented here for the first time are nucleotide-binding parameters in the absence of Mg²⁺ for mutants Phe⁴⁸⁷ Leu, Arg⁴⁸⁹ Leu, and Lys⁴⁹² Tyr of the⁴⁸⁷FSRDRK loop in domain N. These latter mutants have only been analyzed previously in the presence of Mg²⁺ (5). We have shown before (33, 34), and reproduced in Table I, that wild type exhibits a striking Mg²⁺ dependence for both TNP-8N₃-ATP and ATP binding; Mg²⁺ lowers the apparent affinity for the TNP-nucleotide and increases (as much as 40-fold) that for ATP.

View larger version (24K): [in this window] [in a new window]   FIG. 5. Binding of ATP determined by inhibition of TNP-8N3-ATP photolabeling in the presence (closed symbols) or absence (open symbols) of Mg2+ Examples of data for Thr441 Ala (squares), Glu442 Ala (triangles), Arg560 Leu (diamonds), and Leu562 Phe (circles) are shown together with the best fits to the data for the expressed wild type obtained in the presence and absence of Mg2+ (short dashed and long dashed lines, respectively) (5, 33). Photolabeling was performed in 25 mM EPPS/TMAH (pH 8.5), 20% (v/v) glycerol at a [-32P]TNP-8N3-ATP concentration of 3x K0.5 and the indicated concentrations of ATP with either 1 mM MgCl2 and 0.5 mM EGTA (presence of Mg2+) or 2 mM EDTA (absence of Mg2+). The lines show the best fits to the data of the equations described under "Experimental Procedures." Table I lists KD values derived in this way.

Although the binding sites for ATP and TNP-8N₃-ATP may not be identical, there is sufficient overlap to ensure efficient competition when ATP or MgATP is added at varying concentrations during photolabeling, and we have previously demonstrated that the competition assay provides highly accurate values for the ATP and MgATP binding affinities of wild type and mutant Ca²⁺-ATPases expressed in COS-1 cells (5). For wild type, the true affinities (K_D values) for MgATP and ATP calculated under the assumption of competitive inhibition as described previously (5) were found to be 0.51 and 21 µM, respectively (Table I). Fig. 5 shows examples of the experimental data from which the K_D values for ATP and MgATP in Table I were derived, corresponding to Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, Arg⁵⁶⁰ Leu, and Leu⁵⁶² Phe, with data for wild type (see also Refs. 5 and 33) indicated by dashed lines without data points. Because of the deficient photolabeling described above, competition experiments could not be carried out for Lys⁵¹⁵ Ala and Arg⁵⁶⁰ Glu. Mutation Arg⁵⁶⁰ Leu decreased the affinity (i.e. increased K_D) for MgATP more than 1000-fold, whereas mutations Cys⁵⁶¹ Ala and Cys⁵⁶¹ Trp had little effect, as mentioned above. A decrease of affinity for MgATP of 40-fold relative to wild type was seen for Arg⁵⁶⁰ Val, and as much as 70–180-fold for Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Leu⁵⁶² Phe. In the absence of Mg²⁺, Arg⁵⁶⁰ Leu still showed a conspicuous 30-fold decrease of affinity for ATP, whereas for Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Leu⁵⁶² Phe, the affinity for ATP was only 2–5-fold reduced relative to wild type.

In the case of the⁴⁸⁷FSRDRK mutants, Arg⁴⁸⁹ Leu and Lys⁴⁹² Tyr showed extremely poor binding of ATP in the absence of Mg²⁺, reproducing what was found for the TNP-nucleotide. For Phe⁴⁸⁷ Leu the changes in the presence and absence of Mg²⁺ were rather similar, 10–20-fold reduced affinity compared with wild type.

For Thr⁴⁴¹ Ala and Arg⁵⁶⁰ Leu, competition studies were also carried out with ADP. For wild type, the K_D values for ADP in the presence and absence of Mg²⁺ were 4 and 28 µ_M, respectively. For both Thr⁴⁴¹ Ala and Arg⁵⁶⁰ Leu, there were pronounced decreases in affinity for ADP as well as MgADP (8- and >100-fold, respectively, for ADP, and 30- and 200-fold, respectively, for MgADP).

CrATP-induced Ca²⁺ Occlusion—The conspicuous effects found in the above-described experiments testing the nucleotide-binding properties of the mutants, and the uncertainty with respect to Lys⁵¹⁵ Ala, because it could not be photolabeled with TNP-8N₃-ATP, encouraged us to study, as an alternative, the binding of another nucleotide, the non-phosphorylating ,-bidentate chromium(III) complex of ATP ("CrATP"). In the wild type, CrATP binds with relatively low affinity, forming a very stable complex with the enzyme, in which Ca²⁺ is "occluded" at the transmembranous high affinity Ca²⁺ sites (30, 35). The high stability of the Ca²⁺-occluded CrATP-enzyme complex upon removal of free CrATP from the medium (35) makes it feasible to quantify the complex by size-exclusion chromatography of detergent-solubilized microsomal protein following incubation with CrATP in the presence of radioactive ⁴⁵Ca²⁺, and we have previously demonstrated a correlation between deficient CrATP-induced ⁴⁵Ca²⁺ occlusion and defective ATP binding in mutants of the Ca²⁺-ATPase (5, 33). Because the Ca²⁺ titration of ATPase activity described above indicated that all the mutant enzymes bind Ca²⁺ with an affinity similar or close to that of wild type, the Ca²⁺ sites are saturated at the ⁴⁵Ca²⁺ concentration of 40 µ_M used here, and the amount of ⁴⁵Ca²⁺ occluded should, therefore, reflect the ability to bind CrATP. Fig. 6 shows ⁴⁵Ca²⁺ elution profiles obtained for wild type and selected mutants, as well as for control microsomes harvested from cells transfected with the expression vector without insert. For wild type, a distinct peak is seen at 14–15 min of retention time, corresponding to the elution of Ca²⁺-ATPase, whereas the control only shows a rather broad peak extending between 10 and 16 min of retention time. The difference between the test and control curves at 14–15 min of retention time provides a measure of ⁴⁵Ca²⁺ bound in the occluded state in the Ca²⁺-ATPase. For mutant Thr⁴⁴¹ Ala, a minor peak of radioactivity was evident at the position corresponding to Ca²⁺-ATPase, indicating partial, but extremely reduced, formation of the Ca²⁺-occluded CrATP-enzyme complex in the present incubation conditions. No such peak was obtained in experiments performed with mutants Glu⁴⁴² Ala, Lys⁵¹⁵ Ala, Arg⁵⁶⁰ Leu, and Leu⁵⁶² Phe (Fig. 6), or with Arg⁵⁶⁰ Val and Arg⁵⁶⁰ Glu (data not shown). For the Cys⁵⁶¹ mutants, the ⁴⁵Ca²⁺ peak corresponding to Ca²⁺-ATPase was similar to that seen for wild type (data not shown). Thus, the effects of the mutations on formation of the Ca²⁺-occluded CrATP-enzyme complex are in accordance with the effects on MgATP binding described above, and Lys⁵¹⁵ Ala is clearly included among the mutants showing defective nucleotide binding.

View larger version (24K): [in this window] [in a new window]   FIG. 6. CrATP-dependent Ca2+ occlusion. Wild-type or mutant Ca2+-ATPase expressed in COS-1 cell microsomes was incubated for 1 h at 37 °C in 50 mM TES/Tris (pH 7.0), 100 mM NaCl, 5 mM MgCl2, 40 µM 45CaCl2, and 1 mM CrATP. At the end of the incubation period, the membranes were solubilized by addition of 5 mg/ml of the non-ionic detergent C12E8. Following centrifugation, the supernatant was subjected to size-exclusion high pressure liquid chromatography as described previously (5). The elution buffer contained 50 mM TES/Tris (pH 7.0), 100 mM NaCl, 10 mM MgCl2, 1.5 mM CaCl2, 1 mM EGTA, and 2 mg/ml C12E8. The solid circles show the radioactivity in collected fractions. The result of a control experiment with microsomes harvested from cells transfected with the expression vector without insert is indicated by the thin line without data points. The difference between the test and control curves at 14–15 min of retention time, corresponding to the elution of Ca2+-ATPase, provides a measure of the occluded 45Ca2+. The amount of mutant Ca2+-ATPase protein applied to the column was roughly equivalent to the amount of wild type (±30%, due to variation of the expression level, as determined by enzyme-linked immunosorbent assay).

The E₁PCa₂ to E₂P Conformational Transition—To examine the effects of the mutations on the E₁PCa₂ E₂P transition, we determined the rate of decay of phosphoenzyme under conditions (0 °C, presence of K⁺, and neutral pH) where the E₁PCa₂ E₂P transition is rather slow and, thus, rate-limiting for the dephosphorylation, whereas the subsequent hydrolysis of E₂P (Scheme 1) is relatively rapid. The phosphorylation by [-³²P]MgATP was terminated by addition of an excess of EGTA (to remove Ca²⁺). As seen in Fig. 7 and Table II (column marked "EGTA," i.e. no addition of non-labeled MgATP), the rate of phosphoenzyme decay was reduced as much as 6–8-fold relative to wild type, for mutants Thr⁴⁴¹ Ala, Glu⁴⁴² Ala, and Lys⁵¹⁵ Ala. Leu⁵⁶² Phe deviated much less from wild type. Among the Arg⁵⁶⁰ mutants, only Arg⁵⁶⁰ Val could be examined, because of the very low phosphorylation levels obtained with the other two Arg⁵⁶⁰ mutants at 0 °C. It appears from Fig. 7 and Table II that Arg⁵⁶⁰ Val showed a unique 2-fold enhancement of the rate of phosphoenzyme decay, relative to wild type.

View larger version (33K): [in this window] [in a new window]   FIG. 7. Dephosphorylation at 0 °C of phosphoenzyme formed from ATP. Phosphorylation was performed for 15 s at 0 °C in a medium containing 40 mM MOPS/Tris (pH 7.0), 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, 0.955 mM CaCl2 (giving a free Ca2+ concentration during phosphorylation of 10 µM), 10 µM calcium ionophore A23187 [GenBank] , and 5 µM [-32P]ATP (50 µM in case of Arg560 Val, because of its low affinity, cf. Fig. 2A). To measure dephosphorylation, the phosphoenzyme was chased by addition of 10 mM EGTA (solid triangles), 10 mM EGTA with 1 mM non-radioactive MgATP (open circles), or 10 mM EGTA with 5 mM non-radioactive MgATP (solid circles), and acid quenching was performed at the indicated time intervals. The lines show the best fits of a monoexponential decay function; the rate constants are listed in Table II.

View this table: [in this window] [in a new window]   TABLE II Dephosphorylation of phosphoenzyme formed from [-32P]ATP or 32Pi Formation of E1PCa2 from [-32P]ATP at 0 °C or of E2P from 32Pi at 25 °C, and subsequent chase of the phosphoenzyme with dephosphorylation buffer containing excess EGTA (to terminate phosphorylation from [-32P]ATP by removal of Ca2+) or excess EDTA (to terminate phosphorylation from 32Pi by removal of Mg2+) and varying amounts of non-radioactive MgATP or ATP as indicated, was carried out as described in the legends to Figs. 7 and 9. A monoexponential decay function was fitted to the data, and the rate constant (kdephos) is shown. ADP sensitivity was tested on phosphoenzyme formed from [-32P]ATP by addition of 1 mM ADP with 10 mM EGTA (final concentrations), and the fraction of the phosphoenzyme remaining after 5 s ("EP") is shown.

View larger version (28K): [in this window] [in a new window]   FIG. 9. Dephosphorylation of phosphoenzyme formed from Pi. Wild type and mutants were phosphorylated at 25 °C for 10 min in a medium containing 100 mM MES/Tris (pH 6.0), 2 mM EGTA, 30% (v/v) dimethyl sulfoxide, 10 mM MgCl2, and 0.5 mM 32Pi. Dephosphorylation was studied at 25 °C by a 19-fold dilution into a medium containing EDTA (to remove free Mg2+) corresponding to a final concentration of 10 mM, 100 mM MES/Tris (pH 6.0), 2 mM EGTA, 15% (v/v) dimethyl sulfoxide, 0.5 mM non-radioactive Pi, without (open circles) or with (solid circles) 1 mM ATP, followed by acid quenching at serial time intervals. The lines show the best fits of a monoexponential decay function; the rate constants are listed in Table II.

Table II also shows the results of similar measurements performed with mutants Phe⁴⁸⁷ Leu, Arg⁴⁸⁹ Leu, and Lys⁴⁹² Leu, whose MgATP and ATP affinities at the substrate site were reported in the previous paper (5) and in Table I of the present paper, respectively. Lys⁴⁹² Leu showed a conspicuous 6-fold slowing of the phosphoenzyme decay observed upon EGTA addition, whereas only small differences from wild type were seen for Phe⁴⁸⁷ Leu and Arg⁴⁸⁹ Leu. For all 3 mutants, there was a substantial modulatory effect of millimolar MgATP, although slightly less pronounced than for the wild type and the above-described mutants (1.4–1.6-fold enhancement from 1 to 5 mM MgATP).

ADP Sensitivity of the Phosphoenzyme—The two phosphoenzyme intermediates E₁PCa₂ and E₂P can normally be distinguished by their different sensitivity to ADP. E₁PCa₂ is ADP-sensitive, i.e. able to donate the phosphoryl group back to ADP, forming ATP, whereas E₂P is insensitive to ADP and dephosphorylates only by hydrolysis. To test the ADP sensitivity, 1 mM ADP was added following phosphorylation under the same conditions as described for Fig. 7. This resulted in almost complete dephosphorylation of wild type within 5 s, demonstrating accumulation of the E₁PCa₂ form of the phosphoenzyme, and similar effects of ADP were seen for all the mutants studied except Arg⁵⁶⁰ Val, for which as much as 24% phosphoenzyme remained after the 5-s incubation with ADP (Table II), indicating a significantly reduced sensitivity to ADP. This could be the result of accumulation of E₂P, as a consequence of the enhancement of the E₁PCa₂ E₂P transition reported above. Alternatively, the reduced ADP sensitivity reflects a lowered affinity of the E₁PCa₂ form for ADP. This would be in keeping with the very pronounced decrease in affinity for MgADP and ADP shown above for Arg⁵⁶⁰ Leu (Table I).

Properties of the E₂P Phosphoenzyme—The E₂P phosphoenzyme can be formed by "backward" phosphorylation with inorganic phosphate of the E₂ state in the absence of Ca²⁺ and presence of Mg²⁺ (cf. Scheme 1). Fig. 8 shows the phosphorylation at varying concentrations of ³²P_i under conditions optimal for stabilization of the E₂P phosphoenzyme (acidic pH, presence of the organic solvent dimethyl sulfoxide, and absence of alkali metal ions). The concentration of P_i giving half-maximum phosphorylation is close to 10 µM for wild type. Significantly reduced apparent affinity (increased K_0.5) for P_i was seen for Glu⁴⁴² Ala and Lys⁵¹⁵ Ala (5- and 2.5-fold, respectively), whereas Arg⁵⁶⁰ Leu and Arg⁵⁶⁰ Val showed 2-fold increased apparent affinity for P_i. Minor shifts were also seen for mutants Thr⁴⁴¹ Ala and Cys⁵⁶¹ Ala (K_0.5 1.5-fold increased and 1.5-fold decreased, respectively), whereas mutants Arg⁵⁶⁰ Glu and Leu⁵⁶² Phe (Fig. 8), as well as Cys⁵⁶¹ Trp (not shown), were indistinguishable from wild type. It is important to stress here that because the K_0.5 is a function of several kinetic constants, it should not necessarily correlate with the true dissociation constant, K_D, describing the non-covalent enzyme-P_i complex (E₂·P_i in Scheme 1). As discussed previously (34), the respective rate constants, k₂ and k_-2, for formation and hydrolysis of the covalent bond in E₂P may be important as well. To study k_-2, the wild type and selected mutants were phosphorylated under optimal conditions for formation of the E₂P phosphoenzyme, followed by dilution at 25 °C into a medium of the same composition except for the absence of radioactively labeled P_i, a reduction of the dimethyl sulfoxide concentration from 30 to 15%, and the presence of excess EDTA to remove free Mg²⁺ and thus terminate phosphorylation (Fig. 9 and Table II). It should be noted that because of the acidic pH, absence of K⁺, and presence of 15% dimethyl sulfoxide, the rate of dephosphorylation of E₂P is much slower than under physiological conditions. For mutants Glu⁴⁴² Ala and Lys⁵¹⁵ Ala, the reduced apparent affinities for P_i (Fig. 8) were well accounted for by significantly increased rates of E₂P hydrolysis (5- and 3.5-fold, respectively). A similar increase was noted for Lys⁴⁹² Leu (Table II). For mutant Thr⁴⁴¹ Ala, a slightly (1.4-fold) increased rate of E₂P hydrolysis (Fig. 9) explains the 1.5-fold increased K_0.5 (Fig. 8). Leu⁵⁶² Phe, Phe⁴⁸⁷ Ser, Phe⁴⁸⁷ Leu, and Arg⁴⁸⁹ Leu also showed slight increases of the rate of E₂P hydrolysis, and the Cys⁵⁶¹ mutants (data not shown) were indistinguishable from wild type. Interestingly, mutant Arg⁵⁶⁰ Glu showed a conspicuous 5-fold increase of the rate of E₂P hydrolysis (Fig. 9), despite its wild type-like apparent affinity for P_i (Fig. 8). In analogy with this discrepancy, mutants Arg⁵⁶⁰ Leu and Arg⁵⁶⁰ Val showed wild type-like rates of E₂P hydrolysis (Fig. 9), but 2-fold increased apparent affinities for P_i (Fig. 8). Thus, for the Arg⁵⁶⁰ mutants, an increase of the true affinity for P_i (K_D) or an increased rate constant for formation of the covalent bond in E₂P (k₂) must contribute to the change (and lack thereof) in the apparent affinity for P_i.

View larger version (21K): [in this window] [in a new window]   FIG. 8. Phosphorylation from Pi. Wild type and mutants were phosphorylated at 25 °C for 10 min in a medium containing 100 mM MES/Tris (pH 6.0), 2 mM EGTA, 30% (v/v) dimethyl sulfoxide, 10 mM MgCl2, and varying concentrations of 32Pi as indicated. The lines show the best fits of the Hill equation with the Hill coefficient set to 1, giving the K0.5 values (in µM ± S.E.) indicated in parentheses: open circles, wild type (10.5 ± 0.6); open squares, Thr441 Ala (16.4 ± 1.5); open triangles, Glu442 Ala (52.8 ± 8.2); reversed solid triangles, Lys515 Ala (27.5 ± 5.2); open diamonds, Arg560 Leu (4.8 ± 0.4); solid triangles, Arg560 Val (5.0 ± 0.4); solid diamonds, Arg560 Glu (9.0 ± 1.0); reversed open triangles, Cys561 Ala (6.9 ± 0.4); and solid circles, Leu562 Phe (10.6 ± 0.9). The 100% value corresponds to the phosphorylation level reached at infinite Pi concentration as deduced from the fit.

Fig. 9 and Table II furthermore show that inclusion of 1 mM ATP enhanced the dephosphorylation of E₂P close to 2-fold in the wild type and all the mutants studied, thus indicating that the modulatory effect of ATP on this partial reaction step (cf. boxed ATP in Scheme 1) is as unaffected by the mutations as the low affinity modulatory effect on the E₁PCa₂ E₂P transition described above. Because the modulation of E₂P E₂ has been ascribed to metal-free ATP rather than the MgATP complex (14), it is important to note here that when ATP was included in the assay for dephosphorylation of E₂P, the ATP was present in the metal-free form as a consequence of the co-addition of excess EDTA, unlike the assay for the E₁PCa₂ E₂P transition described above, where the major fraction of the ATP was present as MgATP.

Reaction of Lys⁵¹⁵ with FITC—FITC is a potent inhibitor of Ca²⁺-ATPase activity (2) and acts by specific covalent attachment to Lys⁵¹⁵ (3), blocking high affinity ATP binding (2). The specificity of the reaction is mainly due to a high affinity of FITC for the nucleotide site rather than an unusual high reactivity of Lys