Testing the Versatility of the Sarcoplasmic Reticulum Ca2+-ATPase Reaction Cycle When p-Nitrophenyl Phosphate Is the Substrate*

A detailed characterization ofp-nitrophenyl phosphate as energy-donor substrate for the sarcoplasmic reticulum Ca2+-ATPase was undertaken in this study. The fact that p-nitrophenyl phosphate can be hydrolyzed in the presence or absence of Ca2+ by the purified enzyme is consistent with the observed phenomenon of intramolecular uncoupling. Under the most favorable conditions, which include neutral pH, intact microsomal vesicles, and low free Ca2+ in the lumen, the Ca2+/Picoupling ratio was 0.6. A rise or decrease in pH, high free Ca2+ in the lumenal space, or the addition of dimethyl sulfoxide increase the intramolecular uncoupling. Alkaline pH and/or high free Ca2+ in the lumen potentiate the accumulation of enzyme conformations with high Ca2+ affinity. Acidic pH and/or dimethyl sulfoxide favor the accumulation of enzyme conformations with low Ca2+ affinity. Under standard assay conditions, two uncoupled routes, together with a coupled route, are operative during the hydrolysis of p-nitrophenyl phosphate in the presence of Ca2+. The prevalence of any one of the uncoupled catalytic cycles is dependent on the working conditions. The proposed reaction scheme constitutes a general model for understanding the mechanism of intramolecular energy uncoupling.

It is well established that the free energy released from ATP hydrolysis can be used for the vectorial translocation of cations across cellular membranes (1). In this sense, clear structural and functional similarities among P-type ATPases have been observed (2)(3)(4). These similarities include the hydrophobic transmembrane segments containing the porelike region, the extramembranous globular head where the ATP binding site is located, two major conformational states of the enzyme, the acyl phosphate intermediate and the catalytic cycle. Although these common features suggest that ion-transporting ATPases share the same energy transduction mechanism, this idea was challenged when non-nucleotide compounds were used as phosphorylating substrate (5). In fact, it was observed that the transport activity supported by non-nucleotide substrates is dependent on the cation-ATPase involved. In other words, the same non-nucleotide substrate does not always perform the same task since it is dependent on the energy transduction system. For instance, it has been reported that Ca 2ϩ is trans-ported by sarcoplasmic reticulum (SR) 1 Ca 2ϩ -ATPase during the hydrolysis of pNPP or acetyl phosphate (6 -9). However, the hydrolysis of pNPP supports neither Ca 2ϩ transport by the erythrocyte membrane Ca 2ϩ -ATPase (10,11) nor H ϩ transport by the H ϩ -ATPase from the yeast plasma membrane (12). Other studies have reported that reconstituted Na ϩ ,K ϩ -ATPase displays slight Na ϩ uptake during the hydrolysis of acetyl phosphate (13), whereas the H ϩ ,K ϩ -ATPase from gastric mucosa cannot actively transport H ϩ and K ϩ when hydrolyzing acetyl phosphate (5). Thus, it has been recognized that the hydrolysis of non-nucleotide substrates in the absence of transport activity is related to low energy conformations of the enzyme (5,12,14), whereas the simultaneous occurrence of both substrate hydrolysis and ion transport is associated with high energy conformations (7,9,15).
In an attempt to clarify the role of non-nucleotide substrates in the energy coupling process, we selected pNPP as energydonor substrate and SR Ca 2ϩ -ATPase as a transduction system. Initially, the enzyme was exposed to conditions that allowed Ca 2ϩ transport at the expenses of pNPP hydrolysis. Then, the assay conditions were modified to induce uncoupling. The use of a purified enzyme preparation (16) and the highly specific inhibitor TG (17,18) shed light on the hydrolytic activities measured in the presence or absence of Ca 2ϩ . The experimental strategy for analyzing the energy uncoupling routes was based on the inhibitory effect of vanadate and TG, namely vanadate interacts selectively with the E 2 conformation of the enzyme whereas TG inhibits the accumulation of E 1 forms.
Our functional model provides a mechanistic description of the intramolecular uncoupling phenomenon when the SR Ca 2ϩ -ATPase hydrolyzes pNPP. This is relevant for understanding the uncoupling of different P-type ATPases in the presence of different phosphorylating substrates and in the absence of a cationic leak. Sample Preparation-Right side-oriented (intact) vesicles were obtained from the SR membrane of fast twitch rabbit leg muscle as described by Eletr and Inesi (19). Purified Ca 2ϩ -ATPase was prepared by deoxycholate treatment according to method 2 of Meissner et al. (16). Final pellets were resuspended and stored in frozen aliquots at Ϫ80°C until use. Vesicles leaky to Ca 2ϩ were obtained by including A23187 in the reaction medium.

Materials-[
Protein Quantitation-The protein concentration was measured by the Lowry et al. (20) procedure using bovine serum albumin as standard.
Ca 2ϩ Concentration-Free Ca 2ϩ in the external medium was calculated as described by Fabiato (21), taking into account the Ca 2ϩ -EGTA absolute stability constant (22), the pK values for the EGTA protonation (23), pH, and the presence of relevant ligands. Free Ca 2ϩ in the lumenal medium was fixed by equilibrating the vesicles in the standard reaction medium but including potassium oxalate in the range of 0.5-5 mM.
Vanadate Solution and Enzyme Inhibition-Stock solutions containing mostly monovanadate were prepared by dissolving ammonium metavanadate in ultrapure water (Milli-Q grade) adjusted at pH 10.0 with NaOH. The absence of a yellow/orange color indicated the absence of any decavanadate species (24). The sensitivity to vanadate was studied by measuring the dependence of the pNPP hydrolysis rate on the vanadate concentration.
Enzyme Hydrolytic Activity-Linear rates of pNPP hydrolysis were measured at 25°C by following the time-dependent accumulation of p-nitrophenol (25). The standard reaction medium consisted of 20 mM Mops, pH 7.0, 80 mM KCl, 20 mM MgCl 2 , 0.2 mM EGTA, 0.247 mM CaCl 2 (50 M free Ca 2ϩ ), 0.2 mg/ml SR protein, and 5 mM potassium oxalate. The reaction was started by adding a given pNPP concentration. Aliquots containing 0.5 ml of reaction mixture were withdrawn at different time intervals and mixed with 0.5 ml of 10% (w/v) trichloroacetic acid. For each sample, the membrane protein was sedimented at 10,000 rpm and 4°C for 5 min in an Eppendorf microcentrifuge and the supernatant containing p-nitrophenol (0.9 ml) was supplemented with 45 l of 10 N NaOH (final concentration, 0.5 N). A blank assay was performed by adding 0.5 ml of 10% trichloroacetic acid to 0.5 ml of reaction mixture containing no phosphorylating substrate. After mixing, pNPP was added and processed as described before. p-Nitrophenol was quantitated by colorimetric reading at 420 nm. The extinction coefficient of p-nitrophenol (1.62 ϫ 10 4 M Ϫ1 ⅐cm Ϫ1 ) was determined previously under the experimental conditions used in this study. pNPP hydrolysis was also measured under nonstandard conditions. A detailed description of the reaction media composition is provided in the corresponding figure legends.
Transport Experiments-Linear rates of active Ca 2ϩ accumulation were measured at 25°C with the aid of radioactive tracer and sample filtration (26). The standard reaction medium consisted of 20 mM Mops, pH 7.0, 80 mM KCl, 20 mM MgCl 2 , 0.2 mM EGTA, 0.247 mM CaCl 2 (free Ca 2ϩ was 50 M), ϳ7,000 cpm/nmol 45 Ca 2ϩ , 0.2 mg/ml SR protein, and 5 mM potassium oxalate. After equilibration, the reaction was initiated by adding pNPP and stopped at various times by filtering aliquots of 0.5 ml. Filters retaining the microsomal vesicles were rapidly rinsed with 10 ml of ice-cold La 3ϩ medium (10 mM Mops, pH 7.0, 2 mM LaCl 3 ), solubilized, and subjected to liquid scintillation counting. Unspecific 45 Ca 2ϩ retained by the filters was subtracted by performing a blank assay. In this case, an aliquot of 0.5 ml of reaction mixture before the addition of pNPP was filtered and processed as described previously. Other assay conditions for Ca 2ϩ transport are given in the figure captions and in the text where appropriate.
TG Effect on the Purified Ca 2ϩ -ATPase-Hydrolysis of pNPP in the presence or absence of Ca 2ϩ was measured at 25°C, as described for SR vesicles. Measurements in the presence of Ca 2ϩ were carried out in a medium containing 20 mM buffer (Mes, pH 6.0, Mops, pH 7.0, or Tris, pH 8.0), 80 mM KCl, 20 mM MgCl 2 , CaCl 2 and EGTA to yield 50 M free Ca 2ϩ , 0.2 mg/ml purified enzyme, and 10 mM pNPP. For measurements in a nominally Ca 2ϩ -free medium, EGTA was raised to 1 mM and Ca 2ϩ was omitted. Some experiments at neutral pH were performed in the presence of 40% (v/v) Me 2 SO. The reaction medium was supplemented with 2 M TG when indicated. The degree of activation by Ca 2ϩ was calculated as the ratio between the enzyme activity observed in the presence or absence of Ca 2ϩ (both measured in the absence of TG). The TG inhibition factor was calculated as a ratio between the enzyme activity in a Ca 2ϩ -containing medium measured in the absence or presence of TG.
Passive Permeability and Me 2 SO-SR vesicles (0.4 mg/ml) were equilibrated at 25°C in a medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 20 mM MgCl 2 , 0.2 mM EGTA, 0.247 mM CaCl 2 , ϳ9,000 cpm/ nmol 45 Ca 2ϩ , and 5 mM potassium oxalate. The hydrolytic reaction was initiated by adding 10 mM pNPP. The final volume was 2.5 ml and the temperature 25°C. A blank assay was carried out by processing a sample with no added pNPP. At t ϭ 6 min, the reaction medium was supplemented with 5 M TG and a volume of 2.5 ml containing 20 mM Mops, pH 7.0, 80 mM KCl, 20 mM MgCl 2 , 0.2 mM EGTA, 0.247 mM CaCl 2 , 5 mM potassium oxalate, and 80% (v/v) Me 2 SO was added immediately. The time course of Ca 2ϩ accumulation was followed at different times by filtering aliquots containing 0.1 mg of membrane protein. Filters were rinsed with 10 ml of ice-cold La 3ϩ medium, solubilized, and counted by liquid scintillation.
Data Presentation-Experimental data are presented as the mean plus or minus the standard error and correspond to at least three independent assays, each performed in duplicate. Kinetic parameters were evaluated after curve fitting by nonlinear regression algorithm using the SigmaPlot software (Jandel Scientific).

RESULTS
pNPP as a Substrate of the Enzyme-The initial experimental conditions included a preparation of microsomal vesicles showing negligible passive Ca 2ϩ leakage (27), and a reaction medium containing 20 mM Mg 2ϩ . Free Ca 2ϩ in the external medium was maintained at 50 M by including a Ca 2ϩ /EGTAbuffered system while 5 mM oxalate ensured a low free Ca 2ϩ in the lumen. The dependence of the hydrolysis rate on the pNPP concentration measured at pH 7.0 and in the Ca 2ϩ -containing medium displayed a hyperbolic profile, as shown in Fig. 1A. A reaction medium at pH 6.0 or 8.0 did not significantly modify the hydrolysis rate measured at each pNPP concentration.
The dependence of the Ca 2ϩ transport rate on pNPP concentration was also measured. The rate of active Ca 2ϩ accumulation was highly dependent on the reaction medium pH (Fig.  1B). A hyperbolic dependence was observed at pH 6.0 or 7.0, even though higher rates were obtained at neutral pH. Negligible rates were observed after a fast and small component of Ca 2ϩ transport when the experiments were carried out at pH 8.0.
An analysis of the experimental curves (Table I) revealed that K m for pNPP hydrolysis in the presence of Ca 2ϩ and that for Ca 2ϩ transport were in the 2-3 mM range when measured from pH 6.0 to 8.0. Moreover, V max for pNPP hydrolysis in the presence of Ca 2ϩ was around 55 nmol/min/mg of protein, with the pH having only a slight effect whereas V max for Ca 2ϩ transport was clearly affected by pH. The Ca 2ϩ /P i coupling ratio at neutral pH was around 0.6. The coupling ratio decreased to 0.4 when measured at pH 6.0, and complete uncoupling was observed at pH 8.0.
Free Ca 2ϩ in the lumen affects to the enzyme coupling ratio when ATP is the substrate (28,29); therefore, this parameter was also checked. We selected a reaction medium at pH 7.0 containing 20 mM Mg 2ϩ , 50 M free Ca 2ϩ , and 10 mM pNPP that was equilibrated with different concentrations of oxalate. The hydrolysis rate in the presence of Ca 2ϩ displayed a small increase from 42 to 50 nmol/min/mg of protein when the oxalate concentration was raised from 0 to 5 mM ( Fig. 2A). The rate of Ca 2ϩ transport increased as oxalate concentration was raised in the millimolar range, displaying a sigmoidal dependence (Fig. 2B). The rate was negligible in the absence of precipitating anion, whereas maximal values of 31 nmol of Ca 2ϩ /min/mg of protein were obtained in the presence of 5 mM oxalate.
The catalytic properties of the enzyme are sensitive to the presence of Me 2 SO (30, 31); therefore, we evaluated the organic solvent effect in the standard reaction medium at pH 7.0 containing 50 M free Ca 2ϩ , 10 mM pNPP, and native SR vesicles. The pNPP hydrolysis rate in the presence of Ca 2ϩ increased when the organic solvent concentration was raised. Maximal activity was observed at 20% Me 2 SO, whereas a further increase in the solvent concentration up to 40% induced a progressive decrease in the hydrolysis rate (Fig. 3A). In contrast, the hydrolysis rate measured in the absence of Ca 2ϩ increased progressively as the Me 2 SO concentration was raised. Furthermore, maximal Ca 2ϩ transport rates were observed in the presence of 10% Me 2 SO and higher solvent concentrations decreased the transport activity, tending to the zero value (Fig.  3B). The dependence of pNPP hydrolysis in the presence of Ca 2ϩ and that of Ca 2ϩ transport on the Me 2 SO concentration did not show the same profile. Possible effects of Me 2 SO on SR vesicle permeability (i.e. in Ca 2ϩ transport measurements) were ruled out experimentally using 45 Ca 2ϩ -loaded vesicles. An active loading was started by adding 10 mM pNPP to the standard reaction medium and stopped after 6 min by adding 5 M TG. The 45 Ca 2ϩ load was ϳ170 nmol/mg of protein, and a subsequent addition of 40% Me 2 SO had no effect on the Ca 2ϩ content retained by the vesicles (data not shown).
Effect of Vanadate on Enzyme Activity-Vanadate is known to interact with the enzyme conformation in the absence of Ca 2ϩ by acting as a P i analog (32,33). When the reaction medium was at pH 7.0 and contained 20 mM Mg 2ϩ , 1 mM EGTA (i.e. absence of Ca 2ϩ ), 10 mM pNPP, and 5 mM oxalate, the hydrolysis of pNPP by native SR vesicles was completely inhibited by ϳ20 M vanadate (Fig. 4). The inhibition profile in the absence of Ca 2ϩ was exactly the same when a preparation of purified Ca 2ϩ -ATPase was used. Nevertheless, the pNPP hydrolysis rate in a medium containing 50 M free Ca 2ϩ and leaky vesicles required ϳ100 M vanadate to induce complete inhibition. A partial inhibition of 45% was observed in the presence of 100 M vanadate when A23187 was removed (i.e. when intact vesicles were used) and oxalate was included.
Similar experiments using SR vesicles were performed at pH 6.0 (Fig. 5A). In the absence of Ca 2ϩ , complete inhibition of the pNPP hydrolysis was now observed in the presence of ϳ50 M vanadate. Interestingly, the hydrolysis rate in a Ca 2ϩ -containing medium and A23187 (leaky vesicles) displayed the same dependence on vanadate concentration. However, when the enzyme activity was measured in the presence of extravesicular Ca 2ϩ and oxalate (intact vesicles), the sensitivity to vana-   date was lower. A vanadate concentration of 100 M induced a 60% inhibition of the pNPP hydrolysis rate. Fig. 5B shows the sensitivity to vanadate when measured at pH 8.0. The enzymatic activity of SR vesicles in the absence of Ca 2ϩ was completely inhibited by 10 M vanadate. Likewise, the pNPP hydrolysis rate measured in a Ca 2ϩ -containing medium and A23187 (leaky vesicles) was inhibited by 45% in the presence of 100 M vanadate. The percentage of inhibition induced by 100 M vanadate was only 32% when the Ca 2ϩ ionophore was substituted for oxalate (intact vesicles). Fig. 6 shows the effect of lumenal Ca 2ϩ on vanadate sensitivity. The lumenal free Ca 2ϩ was manipulated by modifying the initial concentration of oxalate added (34). Thus, the standard reaction medium at pH 7.0 containing 20 mM Mg 2ϩ and 50 M free Ca 2ϩ in the external medium was equilibrated with either 0.5 or 5 mM oxalate before the addition of 10 mM pNPP. Intact vesicles and an oxalate concentration of 5 mM, i.e. a low free Ca 2ϩ in the lumen, displayed low sensitivity to vanadate. As a reference, 100 M vanadate produced a 45% inhibition of the enzyme activity. When the oxalate concentration was cut to 0.5 mM to increase the free Ca 2ϩ in the lumen, there was a further decrease in vanadate sensitivity. In this case, 100 M vanadate induced a 35% inhibition of the pNPP hydrolysis rate.
The Me 2 SO effect on the enzyme activity was also analyzed by studying vanadate sensitivity (Fig. 7). In one case, the reaction medium at pH 7.0 and in the absence of Ca 2ϩ was supplemented with 40% Me 2 SO and the reaction was started by adding 10 mM pNPP. The inhibition of pNPP hydrolysis in the absence of Ca 2ϩ showed a hyperbolic dependence with respect to vanadate reaching an asymptotic level at ϳ50 M vanadate. In the other case, the reaction medium contained 50 M free Ca 2ϩ , 5 mM oxalate and was also supplemented with 40% Me 2 SO. The inhibition of the enzyme activity in the presence of Ca 2ϩ showed the same vanadate dependence observed in the absence of Ca 2ϩ .
Inhibition of Purified Enzyme by TG-A key question is to know whether or not Ca 2ϩ -ATPase is able to hydrolyze pNPP in the absence of Ca 2ϩ . The hydrolysis of pNPP in the presence or absence of Ca 2ϩ and the effect of TG on these activities were measured using a preparation of purified enzyme. Under these conditions, any hydrolytic activity should be attributed to the Ca 2ϩ -ATPase protein. Fig. 8 (panels A-C) show data obtained at pH 6.0, 7.0, and 8.0, respectively. The rate of pNPP hydrolysis at pH 6.0, in the presence of 10 mM pNPP and in the absence of Ca 2ϩ was 11.0 nmol/min/mg protein. This value decreased to 7.7 nmol/min/mg of protein when measured at pH 7.0 and was 2.7 nmol/min/mg of protein at pH 8.0. The corresponding hydrolytic activities measured in the presence of 50 M free Ca 2ϩ and 10 mM pNPP were 45.7, 62.0, and 42.0 nmol/min/mg of protein, respectively. Therefore, the activating effect of Ca 2ϩ on the pNPP hydrolysis rate increased from 4.2-fold at pH 6.0, to 8.0-fold at pH 7.0 and 15.5-fold at pH 8.0. In any case, the addition of TG to the enzyme in the presence of Ca 2ϩ gave the corresponding activity values measured in the absence of Ca 2ϩ . Moreover, TG had no effect on the pNPP hydrolysis rate measured in the absence of Ca 2ϩ . Data obtained at neutral pH and in the presence of 40% Me 2 SO (Fig. 8, panel  D) show that the enzyme activity in the presence or absence of Ca 2ϩ had the same value and also that TG had the same effect on the rate of pNPP hydrolysis when Ca 2ϩ is present or absent. The activation by Ca 2ϩ or the inhibition by TG calculated as described (see Fig. 8 legend) are dependent on the experimental conditions, increasing from ϳ4 to ϳ15 as the pH increased from 6.0 to 8.0. Neither activation nor inhibition occurred (the factor is 1) in the presence of 40% Me 2 SO. DISCUSSION The coupling efficiency of SR Ca 2ϩ -ATPase during the hydrolysis of pNPP is clearly dependent on pH ( Fig. 1 and Table   FIG I). It is known that Ca 2ϩ binding at equilibrium is a pH-dependent process involving different enzyme conformations and a sequential mechanism (35), as follows.
Therefore, acidic pH stabilizes the enzyme in the absence of Ca 2ϩ (protonated E 2 conformation) and alkaline pH favors Ca 2ϩ binding (E 1 Ca 2 conformation). The activating effect of Ca 2ϩ on the pNPP hydrolysis rate in the purified enzyme is also pH-dependent and increases when the H ϩ concentration decreases (Fig. 8, A-C). This indicates that, under turnover conditions and acidic pH, the protonated E 2 conformations will be more abundant, whereas the E 1 species will predominate at alkaline pH. Qualitatively similar results were obtained when a preparation of intact vesicles and an oxalate-containing medium were used (data not shown).
The highest sensitivity to vanadate was observed during pNPP hydrolysis in the absence of Ca 2ϩ (Figs. 4 and 5). Higher vanadate concentrations were necessary to produce complete inhibition when the H ϩ concentration was raised. This is to be expected since the target species for vanadate, the protonated E 2 forms, will be more abundant at acidic than at alkaline pH. Data obtained with a preparation of purified enzyme (Fig. 4) indicated that pNPP hydrolysis in the absence of Ca 2ϩ is associated with the catalytic activity of the Ca 2ϩ -ATPase protein and not with any other protein.
The effect of vanadate in the presence of Ca 2ϩ was also pH-dependent (Figs. 4 and 5). The hydrolysis of pNPP in leaky vesicles and in the presence of Ca 2ϩ at pH 6.0 was highly sensitive to vanadate. The inhibition profile was the same for leaky vesicles in a Ca 2ϩ -containing medium as for intact vesicles in a Ca 2ϩ -free medium (Fig. 5A). High sensitivity to vanadate is indicative of the predominant accumulation of E 2 forms, especially when pNPP is the substrate. The non-nucleotide substrate does not prevent vanadate binding and inhibition, as occurs with mM ATP (33,36). Moreover, the degree of activation by Ca 2ϩ and inhibition by TG in a preparation of purified enzyme was low (Fig. 8A). TG selectively inhibits the Ca 2ϩ -dependent activity (17), and so high inhibition by TG indicates the predominant accumulation of E 1 forms, whereas low TG inhibition is indicative of a low accumulation of E 1 or what is the same, the predominant presence of E 2 forms.
Enzyme activity in leaky vesicles and in the presence of Ca 2ϩ at pH 8.0 was clearly resistant to vanadate, the inhibition profile being very similar to that displayed by intact vesicles in the presence of Ca 2ϩ and oxalate (Fig. 5B). The Ca 2ϩ activation and TG inhibition factors at pH 8.0 were high (Fig. 8C). These data suggest that E 1 forms will be the predominant species.
Lumenal free Ca 2ϩ has profound effects on coupling when the phosphorylating substrate is ATP (37,38), but not when the phosphorylating substrate is pNPP. The reaction cycle at neutral pH and in the presence of a low lumenal Ca 2ϩ concentration was already highly uncoupled (coupling ratio ϭ 0.6), and the sensitivity to vanadate was relatively low. An increase in internal free Ca 2ϩ produced complete uncoupling and slightly decreased the sensitivity to vanadate. Uncoupling through E 1 forms induced by lumenal Ca 2ϩ has already been characterized when the phosphorylating substrate is UTP (39).
The rate of pNPP hydrolysis had the same value in the presence or absence of Ca 2ϩ when Me 2 SO is 40% (Fig. 3A). In other words, Ca 2ϩ does not activate pNPP hydrolysis in a medium containing 40% Me 2 SO. These are conditions of com- plete uncoupling (Fig. 3B), and we confirmed that Me 2 SO does not alter membrane permeability (data not shown). This lack of Ca 2ϩ activation is associated with a high sensitivity to vanadate, and was the same whether pNPP hydrolysis was measured in the presence or absence of Ca 2ϩ (Fig. 7). In addition, pNPP hydrolysis in the presence of Ca 2ϩ and using purified enzyme was insensitive to TG inhibition (Fig. 8D). These observations can be explained if Me 2 SO favors the reaction cycle in the absence of Ca 2ϩ even when Ca 2ϩ is present. It suggests that one sole catalytic route involving E 2 forms can be forced when the Me 2 SO concentration is high enough.
The activation by Ca 2ϩ and inhibition by TG in a preparation of purified enzyme had different absolute values, but, for each experimental condition tested, the activation and inhibition factors had the same value (Fig. 8). We know that TG produces stabilization of the E 2 conformation (18). Thus, TG inhibits the Ca 2ϩ -dependent activity and allows the hydrolysis of pNPP in the presence of Ca 2ϩ through E 2 forms. This points to the existence of a relationship between the Ca 2ϩ -dependent and Ca 2ϩ -independent activities measured in SR vesicles and suggests that both hydrolytic activities are related to the Ca 2ϩ -ATPase protein.
It has been reported that SR Ca 2ϩ -ATPase is phosphorylated by and/or hydrolyzes ATP (40), UTP (39), or even the nonnucleotide substrate furylacryloylphosphate (41) in a Ca 2ϩ -free medium. In the case of pNPP, the rate of phosphorylation, which is much lower than that by ATP (25), does not allow the accumulation of phosphorylated intermediate during the enzyme turnover. The rates of Ca 2ϩ binding in the presence of pNPP and phosphorylation by pNPP in the absence of Ca 2ϩ must be of similar magnitude in contrast to those observed when ATP is the substrate. This explains why pNPP hydrolysis in a Ca 2ϩ -containing medium can partly occur through E 2 forms.
In conclusion, the low efficiency of Ca 2ϩ transport sustained by pNPP may be attributed to the coexistence of one coupled (E 1 -E 2 cycle) and two uncoupled routes (E 1 cycle and E 2 cycle), When Ca 2ϩ dissociation inside the vesicles takes place before P i release into the external medium, there is an E 1 -E 2 cycle (solid line). This route yields vectorial translocation of Ca 2ϩ from the cytoplasmic to the lumenal compartment. When P i release occurs first, Ca 2ϩ dissociates to the external medium and there is an E 1 cycle. E 2 generated in a coupled cycle can interact with pNPP producing a phosphorylated intermediate without bound Ca 2ϩ (E 2 P) and giving rise to the E 2 cycle. Futile cycles are marked with a dashed line. E 1 Ca 2 and E 1 P⅐Ca 2 are unphosphorylated and phosphorylated conformations, respectively, of the enzyme with high affinity Ca 2ϩ bound. E 2 and E 2 P are unphosphorylated and phosphorylated conformations, respectively, with low Ca 2ϩ affinity. Experiments in a Ca 2ϩ -free media contained 1 mM EGTA and no added Ca 2ϩ . Experiments in the presence of 40% Me 2 SO (D) were performed in the reaction media described for panel B but included the organic solvent. The effect of TG was studied by including a concentration of 2 M. The Ca 2ϩ activation factor was calculated by dividing pNPP hydrolysis obtained in the presence of Ca 2ϩ by that in the absence of Ca 2ϩ when TG was absent. The TG inhibition factor was calculated by dividing the activity in a Ca 2ϩ -containing medium when TG was absent by that when TG was present. The inhibition factor in panel D was corrected taking into consideration the effect of TG on pNPP hydrolysis in the absence of Ca 2ϩ . as depicted in Scheme I. This means that: (i) alkaline pH favors the operation of the E 1 cycle, whereas acidic pH potentiates the E 2 cycle; (ii) uncoupling through the E 1 cycle is favored by the integrity of the vesicles and occurs even when the free Ca 2ϩ in the lumen is low; and (iii) pNPP can be hydrolyzed through the E 2 cycle even in a Ca 2ϩ -containing medium. This study confirms that the catalytic cycle of P-type ATPases does not follow a rigid sequence of reactions and is consistent with the existence of one sole energy transduction mechanism.