(Received for publication, January 3, 1996, and in revised form, October 15, 1996)
From the Departamento de Bioquimica y Biologia Molecular A, Edificio de Veterinaria, Universidad de Murcia, Campus de Espinardo, 30071 Murcia, Spain
Ca2+-ATPase inhibition by stoichiometric and substoichiometric concentrations of cyclopiazonic acid was studied in sarcoplasmic reticulum preparations from rabbit fast-twitch muscle. The apparent affinity of the nonphosphorylated enzyme for ATP showed a Kd of ~3 µM in the absence of cyclopiazonic acid and ~28 µM in the presence of the drug. Fractional saturation of the enzyme by cyclopiazonic acid was accompanied by the appearance of two ATP-binding populations (enzyme with and without drug) and a progressive increase in the half-maximal concentration for saturating the ATP-binding sites. Enzyme turnover in the presence of stoichiometric concentrations of cyclopiazonic acid displayed lower apparent affinity for ATP and lower maximal hydrolytic activity than in the absence of the drug. When cyclopiazonic acid is in the substoichiometric range, the observed kinetic parameters will correspond to the simultaneous contribution of two different reaction cycles sustained by the enzyme with and without drug. The inhibition could be elicited by adding ATP to allow the enzyme turnover when cyclopiazonic acid was preincubated with the enzyme in the presence of Ca2+. The onset of inhibition during enzyme cycling was observed over a period of seconds, revealing the existence of a low inhibition rate constant. It is concluded that cyclopiazonic acid decreases enzyme affinity for ATP in non-turnover conditions by approximately one order of magnitude. This allows enzyme cycling after drug binding, provided that a high ATP concentration is used. Cyclopiazonic acid and ATP do not compete for the same binding site.
Cyclopiazonic acid (CPA)1 is a mycotoxin produced by certain strains of Penicillium cyclopium and Aspergillus flavus. This indole tetramic acid with the molecular formula C20H20N2O3 was first isolated in 1968 (1). It may be found as a natural contaminant in some cereal products and mold-fermented cheese and meat (2), representing a potential risk to human and animal health. The clinical signs of CPA toxicity are usually related to muscle functionality, suggesting a direct effect of CPA on this tissue (3). By using isolated SR vesicles from skeletal muscle, it was proved that CPA is a potent inhibitor of both Ca2+-ATPase activity and ATP-dependent Ca2+ transport (4). The specificity of CPA for SR Ca2+-ATPase and not for other cation ATPases was also established (5). CPA, along with thapsigargin (TG) and 2,5-di(tert-butyl)-1,4-benzohydroquinone, are three structurally unrelated compounds that constitute a group of highly specific inhibitors of SERCA proteins (6).
Selective inhibition by CPA has been exploited in countless reports (for examples, see Refs. 7-12) to analyze Ca2+ signaling mechanisms. Indeed, a detailed characterization of the CPA action is a critical aspect in the manipulation of intracellular Ca2+ stores and may be helpful in providing information on the Ca2+-ATPase energy transduction mechanism. The literature available does not provide a satisfactory description of the CPA effect. For instance, an initial report indicated that the Ca2+-ATPase inhibition was essentially complete at stoichiometric amounts of CPA; however, the type of inhibition was studied in the range of 10-100 nmol CPA/mg of protein (5).
The present study was devoted to clarify the effect of CPA when the drug was used in the stoichiometric range with respect to the enzyme concentration. To this end, a series of measurements on enzyme turnover were performed to elucidate the functional characteristics of the CPA interaction. Ligand binding at equilibrium by using radioactive ATP and TNP-ATP fluorescence allowed us to decipher the relationship between the ATP- and the CPA-binding sites. Measurements of radioactive EP, including different conformational states of the protein, preincubation conditions, and reaction schedules shed light on the enzyme behavior with respect to protection, sensitivity, and onset of inhibition by CPA. All of these data are compiled in a functional model and are discussed with respect to the well characterized inhibitor TG.
[-32P]ATP was obtained from
Amersham Corp. and used at a specific activity of approximately 20,000 cpm/nmol. [3H]Glucose and 45CaCl2
were products of DuPont NEN. They were used at ~3,000 and ~15,000
cpm/nmol, respectively. The Ca2+ ionophore A23187 was
purchased from Boehringer Mannheim. TNP-ATP was from Molecular Probes
Europe, The Netherlands. The liquid scintillation mixture was
Sigma-Fluor (S-4023) from Sigma. A stock solution of
cyclopiazonic acid from Penicillium cyclopium
(Sigma) was prepared in ethanol. The volume of ethanol
added did not exceed 1% of the total volume. CPA was always added to
the incubation/reaction media after the SR vesicles, which were then
preincubated for at least 5 min before the reactions were started.
Sarcoplasmic reticulum vesicles from fast-twitch muscle of rabbit hind leg were prepared according to Eletr and Inesi (13) and stored in frozen aliquots until use. The membrane protein concentration was determined by the colorimetric procedure of Lowry et al. (14) using bovine serum albumin as the standard.
Free Ca2+ ConcentrationThe concentration of free Ca2+ in the EGTA-containing solutions was calculated by the computer program of Fabiato (15) using the binding constant of the Ca2+-EGTA complex (16) and binding constants for EGTA protonation (17).
Ca2+-dependent ATPase ActivityThe steady-state rate of enzyme activity in leaky vesicles was measured at 25 °C. The assay medium was buffered at pH 7.0 and contained Mg2+, K+, Ca2+ ionophore, an ATP-regenerating system, and 10 µM free Ca2+. CPA was also present when indicated. Complete descriptions of the experimental media are given in the corresponding figure captions. The appearance of inorganic phosphate during the first minutes of the reaction was evaluated with molybdovanadate reagent (18). Ca2+-ATPase activity was corrected for Mg2+-dependent ATPase activity measured in the presence of 1 mM EGTA with no added Ca2+.
ATP Binding to the EnzymeNucleotide binding in the absence
of Ca2+ was measured by the double labeling radioactive
technique (19). SR vesicles (0.2 mg protein/ml) were suspended at
22 °C in the incubation medium consisting of 20 mM Mops,
pH 7.0, 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, and specified CPA concentrations. Aliquots of 1 ml
(0.2 mg protein) were layered onto 0.45 µm Millipore filters (HAWP) under vacuum and then rinsed with 1 ml of incubation medium
supplemented with 1-50 µM [-32P]ATP and
1 mM [3H]glucose. The ATP bound to the
protein was estimated from the radioactive counting of the filters
after correction for unspecific nucleotide absorption (3H
labeling).
The fluorescence signal of TNP-ATP was measured with an optical system from Bio-Logic Co. (Claix, France) by using a 150-watt mercury-xenon lamp. The experimental protocol was essentially as described previously (20). The excitation wavelength was 410 nm, and the emission light was selected with a 515-nm cutoff filter (Ealing Electro-Optics, Holliston, MA). Samples were maintained at 22 °C under continuous stirring. The incubation medium contained 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.2 mM EGTA, 0.1 mg/ml SR protein, and 2 µM TNP-ATP in the absence or presence of CPA. The decrease in fluorescence was elicited by successive additions of ATP.
Enzyme Phosphorylation by ATPThe phosphorylated
intermediate of the enzyme was studied by using
[-32P]ATP as a substrate. Unless otherwise stated, all
the reactants were preincubated in an ice-water bath before the
reactions were initiated. The phosphorylation reaction was started by
manual mixing under vortexing of the enzyme suspension with the
radioactive ATP and stopped by the addition of 1 volume of ice-cold
perchloric acid (0.25 M) plus sodium phosphate (2 mM). The quenched samples were filtered under vacuum using
0.45-µm pore size Millipore filters (type HAWP), and the filters were
then washed five times with 5 ml of 0.125 M perchloric acid
and 1 mM sodium phosphate. The filters were subjected to
radioactive counting after solubilization in 3 ml of liquid
scintillation mixture. Different enzymatic forms, reaction media, and
experimental schedules were applied to study the effect of CPA as
follows:
(i) Maximal levels of EP initiated from E1Ca2.
The microsomal vesicles (0.2 mg protein/ml) were suspended in the
presence of 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, 0.967 mM CaCl2 (10 µM free
Ca2+), and 15 µM A23187. This reaction
mixture was preincubated with a defined CPA concentration. The
phosphorylation reaction was started by mixing 0.5 ml of this medium
with 20 µl of medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, 0.967 mM CaCl2, and 1.25 mM [-32P]ATP (50 µM final
concentration). The reaction was quenched after 5-s by an equal volume
(0.5 ml) of quenching solution.
(ii) Maximal levels of EP initiated from E2. SR vesicles
(0.2 mg/ml) were incubated with 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, and 15 µM A23187. Following an
additional incubation with a defined CPA concentration, 0.5-ml aliquots
were mixed with the phosphorylation medium (20 µl) containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM
MgCl2, 24.2 mM CaCl2, and 1.25 mM [-32P]ATP. The final ATP and
Ca2+ concentrations were 50 µM and 0.967 mM, respectively. The reaction was stopped by the addition
of acid after 5 s.
(iii) EP evolution initiated from E1Ca2. The
enzyme suspension containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, 0.967 mM CaCl2, 0.2 mg
protein/ml, and 15 µM A23187 was preincubated with 0.8 µM CPA. CPA was omitted from the control assay. The
phosphorylation medium contained 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, 0.967 mM CaCl2, and 10 mM [-32P]ATP. The reaction was started by
mixing 0.5 ml of enzyme suspension with 20 µl of radioactive ATP
medium (400 µM [
-32P]ATP after mixing)
at 22 °C and arrested in the second time scale by adding 0.5 ml of
acid.
(iv) Chase of maximal EP initiated from E1Ca2.
Pretreatment of SR vesicles (0.2 mg/ml) in a medium of 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM
MgCl2, 1 mM EGTA, 0.967 mM
CaCl2, 15 µM A23187, 2 mM
phosphoenolpyruvate, and 10 units/ml pyruvate kinase with a certain CPA
concentration was carried out at 22 °C. The reaction was studied at
the same temperature by an initial addition of 50 µM
nonradioactive ATP. After 3 min, the reaction mixture was supplemented
with ~500,000 cpm of [-32P]ATP to give approximately
20,000 cpm/nmol after mixing. Samples were denatured 2 s later by
the addition of the perchloric acid/phosphate solution (0.5 ml).
(v) Maximal levels of EP initiated from E2ATP. The initial
incubation medium consisted of 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 1 mM EGTA, 20 or 200 µM
[-32P]ATP, 0.2 mg/ml SR protein, and 15 µM A23187. CPA was then added to the incubation medium.
The reaction was started by mixing 0.5 ml of enzyme suspension with 20 µl of medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, and 24.2 mM CaCl2. The phosphorylation time was 2 s, and the reaction was terminated by acid quenching.
The experimental values represent the average of at least three independent experiments performed in duplicate. The standard deviations of the mean values (plus or minus) are given when indicated.
The inhibition
of SERCA activity is critically dependent on enzyme, ATP, and CPA
concentrations (5); therefore, we initially studied the CPA/enzyme
stoichiometry under our experimental conditions. The steady-state rate
of enzyme activity as a function of SR protein concentration was
measured at 25 °C in leaky vesicles in the presence of an
ATP-regenerating system and a fixed ATP concentration of 50 µM. As can be seen in Fig. 1, the rate of
ATP hydrolysis was linearly dependent on the membrane protein
concentration (). When the experiments were repeated in the presence
of CPA, the linear dependence of the enzyme activity was delayed. This
effect was more evident for a CPA concentration of 0.25 µM (
) than for a concentration of 0.1 µM
(
), which is clear proof of high affinity inhibition. An abscissa
intersection value of 0.025 mg/ml was obtained in the experiments with
0.1 µM CPA and 0.055 mg/ml in the presence of 0.25 µM CPA. The asymptote intersection with the abscissa axis
provides information on the drug/enzyme stoichiometry. Thus, a
drug:enzyme molar ratio of approximately 1:1 can be deduced, assuming
from the maximal EP level that 1 mg of SR protein contains ~4 nmol of
Ca2+-ATPase active sites.
It was published previously that ATP counteracts the inhibitory effect
of CPA (5). Therefore, we studied whether CPA perturbed the interaction
of ATP with the enzyme by measuring the equilibrium binding of ATP to
the nonphosphorylated enzyme (Fig. 2A). The assay was carried out in the presence of EGTA to prevent any ATP hydrolysis. Protein concentration was 0.2 mg/ml, and radioactive ATP
was used as a tracer. The ATP-binding isotherm in the absence of CPA
() indicates a single population of ATP-binding sites with a maximal
binding capacity of ~4 nmol/mg protein and an apparent Kd of ~3 µM, which is in agreement
with previous reports (20, 21). When the experiments were performed in
the presence of 0.3 or 0.8 µM CPA, there was a
progressive increase in the half-maximal ATP concentration necessary
for saturating the high affinity binding sites. It was not possible to
measure the saturation level in the presence of CPA due to unspecific
low affinity ATP binding (21). Nevertheless, the high affinity ATP
sites can be characterized by a Scatchard plot (Fig. 2B).
The titration curve in the presence of 0.3 µM CPA
displayed a biphasic pattern (Fig. 2B, inset), confirming
the existence of two different ATP-binding populations (enzyme with and
without CPA) at substoichiometric drug concentrations. The apparent
Kd for ATP in the presence of CPA (the lower
affinity component) can be more easily evaluated when the enzyme is
saturated by the drug (i.e. with 0.8 µM CPA, equivalent to 1 mol of CPA/mol of enzyme). Under these conditions, we
found a single population of ATP sites with maximal binding of 4 nmol/mg of protein and an apparent Kd of ~ 28 µM (Fig. 2B, main panel).
To complement the preceding data and also to test CPA concentrations
above the stoichiometric level, we took advantage of the fluorescent
probe TNP-ATP (22, 23). The binding of TNP-ATP to the enzyme in the
absence of Ca2+ is accompanied by a specific fluorescence
increase. An indirect titration of the catalytic site can be obtained
by the subsequent sequential addition of ATP because the change in
relative fluorescence can be correlated with nanomoles of ATP bound to
the enzyme (20, 24). The results (Fig. 3) confirmed that
CPA increases the half-maximal concentration necessary for saturating
the ATP-binding sites and also that maximal ATP binding can reach the
level observed in the absence of drug. Moreover, CPA lowered the Hill
coefficient of the ATP-binding curve, when the CPA:enzyme molar ratio
was <1. This is illustrated by comparing the fluorescence isotherm in
the absence of CPA () and that obtained at 0.25 µM CPA
(
). The selected CPA concentration represents approximately 60%
enzyme saturation by the drug because the SR protein concentration was 0.1 mg/ml. We also extended our measurements to 0.4 (
) and 1.2 µM CPA (
), representing 1 or 3 mol of CPA/mol of
enzyme, respectively. It is noteworthy that the half-maximal saturation
in the fluorescence isotherms was independent of CPA when the
CPA/enzyme was
1. Likewise, the ATP binding cooperativity reverted to
a Hill coefficient of approximately 1 when the enzyme was saturated by
CPA.
We also studied the Ca2+-ATPase activity as a function of
the ATP concentration after incubation of the nonphosphorylated enzyme with CPA. The drug concentration was selected to give the same CPA:enzyme molar ratios used in the ATP-binding experiments of (Fig.
2). The reaction medium contained a protein concentration of 0.02 mg/ml
and included Ca2+ ionophore and an ATP-regenerating system
to permit steady-state conditions at ATP concentrations as low as 1 µM. The results of (Fig. 4A)
indicate that samples incubated in the absence of CPA () exhibit the
expected increase of enzyme activity with increasing concentrations of
ATP. The presence of 0.03 µM (
) or 0.08 µM CPA (
) produced a progressive decrease in the
apparent enzyme affinity for ATP and maximal hydrolytic activity. Note
that in the presence of 0.08 µM CPA, a stoichiometric
concentration of the drug, the enzyme may express hydrolytic activity,
provided that the ATP concentration is high enough. It is relevant that CPA concentrations above stoichiometric levels led to further decreases
in enzyme activity (data not shown). Enzyme activity data obtained at
the higher ATP concentrations were also analyzed in a double reciprocal
plot (Fig. 4B). The absence of a common intersection point
on the ordinate axis confirms that the Vmax value is dependent on the CPA concentration.
CPA and Reaction Cycle
It was shown previously by using the
extrinsic fluorescence of fluorescein-5-isothiocyanate that CPA
stabilizes the E2 form of the enzyme (5, 25). This
observation on enzyme conformational states can be more deeply analyzed
by considering the basic molecular transitions involved in the
Ca2+-ATPase reaction cycle (see Scheme I).
Thus, the reactivity of the E1 form toward ATP can be
monitored by following the phosphorylation partial reaction (step 2).
For this purpose, 0.2 mg/ml SR vesicles preequilibrated with saturating
Ca2+ (E1Ca2) were supplemented with
CPA up to 1 µM, and the accumulated EP was studied at
0 °C by measuring the 32P covalently bound to the
protein after addition of [
-32P]ATP. The EP level
reached the same maximal levels as those obtained in the absence of
drug (Fig. 5). Alternatively, SR vesicles preincubated in the presence of excess EGTA (enzyme deprived of Ca2+ or
E2) were incubated with CPA before the addition of
[
-32P]ATP and Ca2+. The phosphorylation
reaction was also performed at 0 °C and maintained for 5 s as
before. Using this protocol, we can study the sequence of steps 1 and
2. In this case, the presence of increasing CPA concentrations produced
a progressive inhibition of the radioactive EP level (Fig.
6).
The time course of the accumulated EP was also considered in the next
assay. The experiment was performed at 22 °C and involved the
addition of CPA to the enzyme in the presence of Ca2+,
before mixing with radioactive ATP. The reaction was stopped at various
time intervals by the addition of acid. The ATP concentration was
relatively high (400 µM) to permit observation of the EP
time course on the scale of minutes. This also permitted detection of
enzyme turnover in the presence of 0.8 µM CPA (a
stoichiometric concentration). As illustrated by (Fig.
7), CPA inhibition can be elicited by allowing enzyme
turnover. The EP hydrolytic cleavage was monophasic until complete
exhaustion of the substrate in the absence of CPA. However, when CPA
was present, there was a clear biphasic response. Enzyme inhibition
manifested as an initial transient decay of EP followed by a lasting
steady-state level. As a result of the inhibition, the ATP was not
completely consumed in the same time span as in the control experiment.
The different temperature used may also explain why the EP level
measured at the shorter phosphorylation times was lower than the
control value obtained in Fig. 6. We then studied inhibition of the
cycling enzyme as a function of different CPA concentrations. The
reaction medium was modified to maintain enzyme turnover in the minute time scale without the necessity of adding high concentrations of
radioactive ATP. CPA was added to the enzyme in the presence of
Ca2+, as before, but the phosphorylating substrate was now
50 µM nonradioactive ATP. The reaction mixture also
contained an ATP-regenerating system to ensure enzyme turnover during
the assay. After a phosphorylation period of 3 min at 22 °C, EP
labeling was detected by adding a pulse of [-32P]ATP
followed by acid quenching at 2 s. Fig. 8 shows
that the accumulated EP under steady-state conditions decreased as a
function of the CPA concentration. The degree of inhibition at 0.8 µM CPA was higher than in the experiment of Fig. 7
because of the presence of a lower ATP concentration.
The last set of experiments was directed toward establishing whether
ATP had any protective effect when CPA was added to the enzyme in the
E2 form. In the experiment outlined in Fig.
9, we exposed SR vesicles in the absence of
Ca2+ to 20 or 200 µM
[-32P]ATP before the addition of different CPA
concentrations in the stoichiometric range. Then, Ca2+ was
added to initiate enzyme phosphorylation. The reaction was carried out
at the ice-water temperature for 2 s. The observed inhibition
resembles that obtained when CPA was directly added to the enzyme in
the absence of ATP and Ca2+ (compare Fig. 6). EP decreased
progressively as the CPA concentration rose. However, for a given CPA
concentration, the EP level was higher in the experiment performed in
the presence of the higher ATP concentration.
The experiments reported above are helpful for understanding the functional properties of CPA as inhibitor of the SERCA family of proteins. It should be stressed that the present study has been restricted to stoichiometric and substoichiometric CPA concentrations. Because different CPA:enzyme molar ratios have been reported (5, 25) as producing complete inhibition of the Ca2+-ATPase activity, our first concern was to re-examine this parameter under well defined assay conditions. A plot of enzyme activity versus membrane protein concentration (Fig. 1), or the inhibitory effect of CPA on EP formation (Fig. 6), indicated a 1:1 stoichiometry with respect to active sites on the enzyme. Therefore, the number of CPA-binding sites is equivalent to that of the high affinity ATP-binding sites. The stoichiometric value is somewhat lower than that originally reported (6-8 nmol CPA/mg of protein) at low ATP concentrations (5). Small changes in the experimental conditions, including the purity of the membrane protein preparation or the methods used for the binding site estimation, may account for the observed difference.
When the enzyme is completely saturated by CPA, the Scatchard plot
(Fig. 2B) indicates the existence of a single ATP-binding population. This is consistent with the binding cooperativity data
(Fig. 3) showing a Hill coefficient of approximately 1. However, when
the enzyme is partially saturated by CPA, there are two nucleotide binding populations (Fig. 2B, inset) and negative
cooperativity (Fig. 3). The apparent Kd for ATP in
the presence of CPA can be graphically estimated from a Scatchard plot,
and the evaluation is easier when the enzyme is completely saturated by CPA (Fig. 2B, compare data points in main panel
and low affinity component in the inset). Furthermore,
the half-maximal ATP concentration necessary for saturating the high
affinity sites will only give information on the apparent
Kd when CPA:enzyme is 1. In any case, CPA reduced
the apparent Kd for ATP by approximately one order
of magnitude. This may explain why other phosphorylating substrates
such as acetylphosphate, which has lower affinity for the enzyme,
cannot be hydrolyzed after incubation with CPA (5). Interestingly, the
effect of TG as a SERCA inhibitor is similar because TG also decreases
the binding affinity of ATP (20). This effect may explain the observed
ATP binding inhibition induced by TG (26). A decrease in the binding
affinity at the phosphorylation site promoted by TG would also account
for the loss in the phosphorylating capacity of Pi (27,
28).
Earlier studies have claimed a competitive relationship between CPA and ATP (5, 25). However, our ATP-binding experiments indicate that the observed decrease in the Kd for ATP is independent of CPA. Certainly, this observation cannot be reconciled with a direct competition between CPA and ATP. Furthermore, CPA does not modify the TNP-ATP fluorescence signal, suggesting that the ligands do not interact at the same binding site. This property is shared by TG. Studies on TNP-ATP fluorescence suggested that the interaction of TG with the protein was not at the ATP-binding site (20).
The reversal of CPA inhibition by increasing the ATP concentration has been interpreted as a protective effect of ATP and attributed to a competitive effect (5). From the present data (Figs. 2, 3, 4), it is apparent that CPA simply decreases the Kd for ATP so that the enzyme will express higher hydrolytic activity if ATP concentration is increased. A similar effect of ATP was also observed in the EP measurements of Fig. 9. Before it can be assured that CPA is a competitive inhibitor, it must be demonstrated that the Ca2+-ATPase activity in the presence of CPA can reach the maximal velocity measured in the absence of drug. A direct plot of enzyme activity versus ATP concentration (Fig. 4A) indicates that this is not the case. Double reciprocal plots of data conforming a high affinity inhibition mechanism give concave-down curves that are difficult to analyze (29). However, by plotting enzyme activity data obtained at the higher ATP concentrations, we were able to analyze the region near the 1/v axis in more detail. Fig. 4B confirms that CPA does not behave as a competitive inhibitor. The conclusion that CPA is a competitive inhibitor (5) was based on the analysis of nonlinear double reciprocal plots showing a poor resolution in the region near the ordinate axis. Furthermore, the experiments were performed at CPA concentrations (10-100 nmol/mg of protein), much higher than the stoichiometric level. In another study (25), a Dixon plot was the only experimental evidence offered to support the competitive character of CPA. This plot provides information on the Ki value but cannot be used to diagnose the type of inhibition.
The Ca2+-ATPase activity isotherms shown in Fig.
4A indicate that the enzyme may undergo turnover in the
presence of stoichiometric CPA at high ATP concentrations. However,
enzyme turnover displayed lower ATP affinity
(KATP) and lower maximal velocity
(V
max) than in the absence of CPA.
Consequently, when the CPA concentration is in the substoichiometric
level, the observed kinetic parameters will correspond to the
simultaneous contribution of two different reaction cycles sustained by
the enzyme without CPA (KATP and Vmax) and the enzyme with CPA
(K
ATP and V
max). The
inhibition of the SERCA activity by CPA can be illustrated by Scheme
II. In this model, CPA does not interact with the
E1 form, and the binding of CPA to E2 does not
take place at the catalytic site. When CPA was in excess with respect
to the enzyme protein concentration, there was a higher degree of
inhibition. This suggests that the CPA effect is not restricted to the
ATP-binding process, although this was not investigated further.
The phosphorylation of E1Ca2 by ATP was not altered by CPA (Fig. 5); hence, the catalytic site is probably not the drug target. This suggests that E1 is refractory to CPA. However, when the reaction cycle was started from E2, the addition of CPA inhibited EP formation (Fig. 6), suggesting a selective effect of CPA on the E2 form. This is in agreement with previous fluorescence measurements of the protein conformational states (5). The protective effect of Ca2+ on E1 can be eliminated by adding ATP. This is an indication that the enzyme turnover generates an intermediate species that is susceptible to CPA. No inhibition was observed in the presence of 50 µM ATP because EP decay took place over a short time span (data not shown). However, the inhibition becomes clear when the EP evolution was studied in the presence of a relatively high ATP concentration (400 µM) to allow enzyme turnover on a time scale of minutes (Fig. 7). The data of Figs. 7 and 8 reveal that the onset of inhibition during the enzyme turnover is a rather slow process. This implies that a high number of reaction cycles is required for enzyme inhibition to be observed. It can be postulated that after the initial binding of CPA to the enzyme, there is a slow isomerization step leading to an E2 state with a lower ATP binding affinity. The data for TG indicated that enzyme inhibition is also associated with the E2 conformation (28). TG inhibition by adding ATP also developed over seconds (30), although it was faster than in the case of CPA inhibition.
An important difference between CPA and TG is that the latter causes a drastic decrease in the apparent Kd for both ATP and Ca2+, producing for practical purposes a dead-end complex (26-28). The present data indicate that CPA diminishes ATP binding affinity but to a lesser extent than TG. Therefore, the enzyme with CPA bound can be forced to undergo turnover by increasing ATP concentration.
Location of the CPA-binding site will throw light on the role played by the region of the SERCA proteins that is recognized by the drug. This may be achieved from the construction of chimeric proteins as was done for the location of the TG target site (31).
We are grateful to Dr. Garcia-Carmona from our Department for helpful suggestions. We also thank the anonymous referee for very positive criticism.