Effect of Diethylstilbestrol and Related Compounds on the Ca2 ...

Vol. 267, No. 17, Issue of June 15, pp. 11923-11929,1992 Printed in U.S. A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Effect of Diethylstilbestrol and Related Compounds on the Ca2'transporting ATPase of Sarcoplasmic Reticulum* (Received for publication, November 22, 1991)

Francisco Martinez-Azorin, Jose A. Teruel, Francisco Fernandez-BeldaS, and Juan C. Gomez-Fernandez From the Departamento de Bwquimica y Biologia Molecular, Facultad de Veterinaria, Uniuersidud de Murcia, 30071 Espinnrdo, Murcia, Spain

Diethylstilbestrol is a potent inhibitory agent of the Ca2+-ATPaseactivity of sarcoplasmic reticulum membranes. Other structurally related molecules, such as dienestrol orhexestrol having hydroxyl groups at para positions of the two benzene rings produce similar effects. The absence or derivatization of the hydroxyl groups as occurs with trans-stilbene or diethylstilbestrol dipropionate converts thestructure in an activating agentof the enzyme. The Ca2+transport profiles in the presence of the referred drugs reproduces the same behavior observed for thehydrolytic activity. There is also a clear indication of a membrane-mediated mechanism of these drugs. Ligand binding experiments at equilibrium indicate that diethylstilbestrol decreases the affinity for Ca2+ of the high affinity Ca2+ sites. Functional studies revealthat theactivation/inhibition induced by these drugs is correlated with decreased levels of phosphoenzyme at steady state, and these levels are sensitive to the Ca2+concentration. Chase experiments of [32P]phosphoenzymeand 4eCaz+indicate a slight activation effect of diethylstilbestrol dipropionate on Caz+ dissociation during the enzyme turnover. The use of different anthroyloxy derivatives of stearic acid as a fluorescent probe suggest that diethylstilbestrol and other inhibitory agents could be located close to the polar region of the lipid bilayer, which interferes with theCa2+-bindingsites, whereas the activators trans-stilbene anddiethylstilbestrol dipropionate may have a deeper position into the membrane, which accelerates the Ca" translocation process.

Diethylstilbestrol (DES)' is an artificial estrogen first synthesized in 1938 (1).In theensuing years, this compound has been widely used as a therapeutic agent to treat different gynecological problems (2) and as a growth promotant in livestock (3). Apart from the use in the referred to cases, which is a matter of much controversy (4,5), DES is a unique probe at the molecular level since it is able to interfere with energy transduction processes catalyzed by H+-ATPases of

* This work was supported by Grant PB 87-0704 from Comision Interministerial de Ciencia y Tecnologia, Spain. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisernent" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ To whom correspondence should be addressed. The abbreviations used are: DES, diethylstilbestrol; n-AS, n-(9anthroy1oxy)stearic acid; EGTA, [ethylenebis(oxyethylenenitrilo)] tetraacetic acid; EP, phosphorylated intermediate of the Ca*'-ATPase; Mops, 4-morpholinepropanesulfonic acid; SR, sarcoplasmic reticulum.

the P-, V-, and F-types (6). Studies on H+-ATPase from rat liver mitochondria (7, 8) have revealed that DES is a potent inhibitor of the whole hydrolytic activity, and more specifically, it has been shown to block the proton translocation activity by actingthrough the Fo sector. In this sense, it evokes the action of dicyclohexylcarbodiimide,although the interaction site appears to be different (7). This knowledge enticed us to study the effect of DES on the energy transduction system associated to the SR membrane, i.e. the Ca'+-transporting ATPase, searching for some common features with other ion motive ATPases. Our results show the modulating effect of DES on the Ca2+ATPase activity underlining the importance of the hydroxyl groups to display an activating or inhibitory response. Likewise, the complex effect of diethylstilbestrol and other molecules was analyzed on the basis of relevant binding sites and the reaction cycle of the enzyme. Measurements with the nAS fluorescence probes supply some clues to the location of DES and related compounds in relation to their modulating capacity. It is concluded that theeffects of these drugs are on the slow conformational transitions associated to Ca'+ binding and dissociation from the enzyme. EXPERIMENTALPROCEDURES

Membrane Preparation-A microsomal fraction enriched in Ca2+ATPase protein was isolated from rabbit leg white muscle as described by Eletr & Inesi (9). The proteincontent was estimated by the procedure of Lowry et al. (10) using bovine serum albumin as standard. Free Ca2+Concentration-Reaction media with different free Ca2+ concentrations were prepared with the aid of a Ca*'-EGTA buffer according to a computer program (11)that takes into consideration the totalconcentrations of Ca2+, EGTA, H+, K', and M e , the stability constant for the Ca2+-EGTAcomplex (12), and the pH dependence of the ionized EGTA species (13). ATPase Activity-The rate of ATP hydrolysis was measured at 25 "C following the liberation of inorganic phosphate with the molybdovanadate reaction (14). A typical experiment contained 20 mM Mops (pH 6.8), 80 mMKC1, 5 mM MgC12,0.05 mg of SR protein/ml, 10 p M A23187,l mM ATP, 1 mM EGTA, and different Caz+concentrations toyield the desired free concentration. Ca2' Transport-The active transport of Ca2+ was measured at room temperature by the radioactive tracer method. The incubation medium included 20 mM Mops (pH 6.8), 80 mM KCl, 10 mM MgC12, 1 mM EGTA, 0.695 mM '5CaC12 (-4000 cpm/nmol), 0.05 mg of protein/ml,5 mM potassium oxalate, 1 mM ATP, and different concentrations of DES or DES dipropionate. The Ca" accumulated inside the vesicles was determined by filtering aliquots of 1 ml (0.05 mg of protein) on HAWP Millipore filters (0.45 pm) at serial time intervals. The filters were rinsed with 4 ml of medium containing 20 mM Mops (pH 6.8), 80 mM KCl, and 1 mM LaC13before counting the 45Ca2+retained on the filters. The rates of Ca*+ transport were calculated from the initial phase of time course plots. Cu2+Binding-The high affinity Ca2' binding was measured by the double-labeling filtration technique (15). The enzyme (0.3 mg of SR/

11923

11924

Diethylstilbestrol and Ca2+-transportingATPase

ml) was incubated at room temperature in amedium consisting of 20 mM Mops (pH 6.8), 80 mM KCI, 5 mM MgCl,, 100 p~ 45CaC12 (-5000 cpm/nmol), 1 mM [3H]glucose(-1000 cpm/nmol), and various EGTA concentrations to obtain different free Ca2+concentrations. In some cases, the incubation medium was supplemented with 20 p~ DES or DES dipropionate. After equilibration for 15-20 min, aliquots of 0.2 ml (0.06 mgof protein) were placed onto filters (Millipore HAWP 0.45 pm) previously soaked in unlabeled medium and subjected to vacuum. Counting of radioactive tracers in the incubation medium and thefilters allowed us to determine the Ca2+bound by subtracting the nonspecific free Ca2+trapped in each filter (3H labeling). Ca2+Dependence on the Intrinsic Protein Fluorescence-Tryptophan fluorescence changes of the Ca2+-ATPasewere measured under stirring at room temperature with a Bio-Logic Optical System (Claix, France). The excitation wavelength (290 nm) was selected by a monochromator, whereas the emission intensity was detected at 90" through a WG320 cut-off filter (Ealing Electro-optics, Holliston, MA). The incubation medium contained 20 mM Mops (pH 6.8), 80 mMKC1, 5 mM MgCl,, 1 mM EGTA, and 0.1 mg of SR protein/ml, The Ca2+titration was performed by adding different Ca2+concentrations andmeasuring the increase in the intrinsic proteinfluorescence. DES was added to theincubation medium prior to theCaz+additions. The fluorescence increases are expressed in relative units, and the total fluorescence changes were calibrated by direct measurement of binding with radioactive tracers (45Ca2+/3H). Phosphorylation by ATP-The phosphoenzyme was obtained in an ice bath by the addition of different concentrations of [Y-~'P]ATP (-10,000 cpm/nmol) to SR vesicles (0.2 mg of protein/ml) suspended in a medium containing 20 mM Mops (pH 6.8), 80 mM KCI, 5 mM MgCI,, 1 mM EGTA, 10 p~ A23187, and either 0.695 mM Ca2+(pCa 5.6) or 1.99 mM Ca2+ (pCa 3). DES or DES dipropionate at a concentration 20 p~ was included in some experiments. The reaction was stopped after 3 s by adding an ice-cold solution containing perchloric acid and sodium phosphate to a final concentrationof 125 and 2 mM, respectively. The quenched samples were centrifuged at 2000 X g for 5 min (4 "C) and processed as previously described (16). Chase of EP Decomposition-32P-Labeled phosphoenzyme formation was carried out for 3 s in an ice-cold solution containing 20mM Mops (pH 6.8), 80 mM KCl, 5 mM MgCl,, 1 mM EGTA, 0.695 mM CaCl,, 10 p M A23187, 0.2 mg of SR protein/ml and 25 p M [r-3'p] ATP (-9000 cpm/nmol). The reaction mixture (0.4 ml) was then diluted with 5 mlof the same medium containing nonradioactive ATP. Labeled phosphoenzyme decay was measured after sequential acid quenching (ice-cold perchloric acid and sodium phosphate) and filtration (0.08 mg). An extensive washing of the HAWP filters with 5 mlof a medium containing 125 mM perchloric acid plus 2 mM sodium phosphate (&fold) preceded the radioactive counting. The effect of DES or DES dipropionate was studied by including a concentration of20pM in the phosphorylation medium. Ca2+Dissociation during the Enzyme Turnover-The phosphorylation reaction was started at room temperature by adding 1mM ATP to a medium of20mM Mops (pH 6.8), 80 mM KCI, 5 mMMgCI2, 0.1 mM "CaCI2 (-10,000 cpm/nmol), 0.1 mM EGTA, 1 mM [3H]glucose (-2000 cpm/nmol), 10 p~ ,423187, and 0.05 mg of protein/ml. After 20 s, aliquots of 2 ml (0.1 mg)were rapidly filtered on HAWP Millipore filters placed in a rapid filtration apparatus and flushed for a controlled period of time with the phosphorylation medium containing nonradioactive Ca2+and glucose. "Caz+and 3Hlabels retained by the filters were counted and used to determine the real amount of Ca2+bound to the enzyme. 20 p~ DES or DES dipropionate were included in the incubation and flushing media when indicated. Drug Location into the Membrane as Monitored by n-AS Flumescent Probes-9-Anthroyloxy derivatives of stearic acid (2-AS, 7-AS, 9-AS, or 12-AS) at a concentration 2 p M were added to a medium of 20 mM Mops (pH 6.8), 80 mM KCI, 10 pg of SR protein/ml, and the suspension was kept for 1-3 h at 30 "C in the dark. Fluorescence intensities were measured with a modular optical system from BioLogic Co. (Claix, France). Samples, under stirring, were excited at 370 nm, and thefluorescence emission was passed through a GG 400 cut-off filter (Ealing Electro-optics, Holliston, MA). A stock solution of lipophilic compounds were prepared inethanol, and the total volume added to the cuvette was always lower than 2%. Correction for the ethanol effect was performed when required. Blanks were run in the absence of a fluorophore. When the fluorescence change was time-dependent, we waited until the transition was completed.The chemical structures of the drugs used in this study are presented in Fig. 1.

DIETHYLSTILBESTROL

DES MPROPIONATE

DIENESTROL

TRANS-STILBENE

HEXESTROL

FIG. 1. Molecular structure of diethylstilbestrol and other

related compounds.

0

20

40

60

D r u g Concentration bY)

c

ue a

0

10

20

D r u g Concentration

60

bY)

FIG. 2. Effect of DES and other counterparts on the Ca2+ATPase activity of SR vesicles. The reaction medium contained 20 mM Mops (pH 6.8), 80 mM KCI, 5 mM MgCl,, 0.05 mgof SR protein/ml, 10 FM A23187, 1 mM EGTA, 0.695 mM CaCl,, and different concentrations of drugs. The reaction was initiated at 25 "C by adding 1 mM ATP and arrested by addition of the phosphate reagent (14). A, effect of DES (0),dienestrol (O), and hexestrol (V); B , effect of DES dipropionate (0)and trans-stilbene (0).Each plotted point represents the initial rate of ATP hydrolysis measured during the first minutes. RESULTS

The Hydrolytic and TransportActiuities-Theeffect of DES and related compounds was initially assessed by measuring the hydrolyticcapacityof the Ca2+-transportingATPase. Thus, SR vesicles in the presenceof the Ca2+ionophore A23187 and increasing concentrations of DES were preincubated under optimal conditions to measure the Ca2+-dependent ATPase activity. The effect elicited by DES was dependent on the concentration used. As depicted in Fig.2 A , it gives rise to a slight but consistent activating effect at low concentrations, whereas higher concentrations exhibit an inhibitory action. When dienestrolor hexestrol was used instead of DES, the experimental curves were similarly shaped even though the

Diethylstilbestrol and Ca2+-transportingATPase activating effect was a little more pronounced and they were shifted on the abscissa axis. These data provide half-inhibition values of approximately 15p~ for DES anda little higher for dienestrol and hexestrol. On the other hand, when DES dipropionate or trans-stilbene were included in the assay medium, they behaved only as activators of the Ca2+-ATPase activity (Fig. 2B). Micromolar concentrations of these compounds activate by the same extent(-40%), with 2.5 pM DES dipropionate and 5 pM trans-stilbene producing half-maximal activation. The distinct effect of these compounds on the enzymatic activity was also reflected on the Ca2+-transporting activity. Ca2+ is actively accumulated by SR vesicles when ATP is added as a phosphate donor substrate. Therefore, we measured thedependence of the net Ca2+entry in the presence of oxalate on the DES concentration added. Fig. 3A shows that the initial rate of Ca2+transport as a function of the DES concentration follows an inhibition pattern similar to that observed for the hydrolytic activity. The Ca2+transport activity was also assayed by including DES dipropionate as a reagent. This compound produces an activating effect on the Ca2+ transport rate (Fig. 3 B ) , matching the corresponding experiments on enzyme turnover. It is then apparent that DES and DES dipropionate canbe considered as representative of the two different responses observed; therefore, we decided to evaluate the influence of these compounds on the Ca2+ dependence of the ATPase

11925

activity. The presence of DES, at concentrations of 10-50 pM (Fig. a), reduces the affinity of the enzyme for Ca2+;therefore, higher Ca2+ concentrations are needed to express the same enzymatic activity. A further increase in the Ca2+concentration displays an inhibitory effect as occurs in the experiments performed in the absence of DES (back inhibition). This is clearly in contrast to the pattern promoted by DES dipropionate (Fig. 4B),where the affinity for Ca2+was unaffected and the maximal hydrolytic activity was increased as the drug concentration was raised. Another feature of interest was that the half-maximal inhibition of the Ca2+-ATPase activity induced by DES was shifted to higher values when the membrane protein concentration rose; i e . , higher concentrations of DES areneeded to reach the same extent of inhibition when the membrane protein increases (Fig. 5). The Ca2+ Transport Domain-In a different set of experiments, we tried to ascertain the effect of DES and analog structures on specific binding sites. Thus, we investigated the Ca2+transport sites by measuring the binding of Ca2+to the Ca2+-ATPaseprotein under equilibrium conditions. In our experiments, the 45Ca2+ concentration was varied in the range of 0.05-25 pM, and the membrane protein concentration was 0.3 mg/ml. The incubation medium, buffered at pH 6.8, was supplemented with 1 mM [3H]glucosein order to evaluate the Ca2+actually bound to theprotein after filtration. A sigmoidal dependence with respect to the Ca2+concentration was observed for the Ca2+binding to thehigh affinity sites (Fig. 6A) as previously published (16). The Ca2+binding experiment was then repeated in the presence of DES or DES dipropio-

n

100

n

Y

B

3

50

j 4

DE8 hY)

0

150 n

n

g

W

100

4 10

50

"

0

10

20

80

DE8 Dipropionate GY) FIG. 3. Effect of DES or DES dipropionate on the initial rate of Ca2+transport. ea2+accumulation was measured at room temperature by the addition of 1 mM ATP to a medium containing 20 mM Mops (pH 6.8),80 mM KCl, 10 mM MgC12,0.695 mM"CaC12, 1 mM EGTA, 5 mM potassium oxalate, and 0.05 mgof SR protein/ ml in the presence of different concentrations of DES ( p a n e l A ) or DES dipropionate (panel B ) . Aliquots of 1 ml were filtered, and the filters were rinsed once with 4 ml of medium 20 mM Mops (pH 6.8), 80 m M KCl, and 1 mM Lacla and subjected to radioactive counting. The reaction was stopped at different time intervals during the first minutes to determine initial rates.

0 8

4

8

2

PC. FIG. 4. Ca2+dependence on the ATPase activity measured in the presence of DES or DES dipropionate. The reaction medium was as described in the legend for Fig.2 containing different ea2+concentrations to yield the required pea. A , the reaction medium was supplemented with DES at the following concentrations: 10 PM (A), 20 p M (W), 30 p M (01,or 50 p M (A).B, the effect of DES dipropionate at the following concentrations: 4 p M (0)or 20 PM (A). The control curve (absence of drug) is represented in both panels by closed circles.

11926

Diethylstilbestrol and Ca'+-transporting ATPase

in the presence of different Ca2+ concentrations. This method provides a very sensitive signal and allowed us to use a low G 100 Y membrane protein concentration(0.1 mg/ml). This procedure h was useful to show a clear effect of DES on Ca2+binding. The fluorescence transition measured in Fig. 6B was modified in amplitude and affinity under the presence of increasing conso centrations of DES (10-30 p ~ ) . The Catalytic and Transport Cycle-A more detailed analysis can be performed by studying the partialreactions involved 4 in thecatalytic mechanism of the Ca2+transport. Addition of 0 [y3'P]ATP to SR vesicles preincubated with Ca2+leads to a 0 20 40 rapidformation of radioactive phosphoenzyme reaching a DES bY) steady-state level that is maintained during the hydrolysis of ATP. Therefore, we canapproach the study of the phosFIG. 5. Membrane protein dependence of the modulating effect of DES on the Ca2+-ATPase activity. The hydrolytic phorylation reaction by measuring the maximal amount of activity was assayed in the presence of different DES concentrations E P formed at steady state once the hydrolytic cleavage of the under the conditions described for Fig. 2. The protein concentrations phosphoprotein is slowed down. This was done in the experwere 0.01 mg/ml (O), 0.05 mg/ml (O), or 0.1 mg/ml (A). iments shown in Fig. 7A, where the phosphorylation reaction was started on iceby adding differentconcentrations of radioactive ATP andstopped 3 s later by acid addition. When the reaction medium contained alow Ca2+concentration (pCa 5.6),the steady-state level of phosphoenzyme increases as a function of the ATP concentration and saturation of the phosphorylating sites (-4 nmol/mg of protein) was attained at a concentration somewhat higher than 10 p ~ . When SR vesicles were preincubated with either DES or

3

E

j

7

6

I

PC.

7

6

5

PC. FIG. 6. Influence of DES or DES dipropionate on the high affinity Ca2+binding to the ATPase protein. A, Ca2+binding was measured by first equilibrating SR vesicles (0.3 mg/ml) in a medium of 20 mM Mops (pH 6.8), 80 mM KC1,5 mMMgC12, 100 pM 46CaC12,1 mM [3H]glucose,different concentrations of EGTA, and either 20 pM DES (A)or 20 p~ DES dipropionate (0).Control values (0)were obtained in the absence of drug. Bound radioactive Ca2+was evaluated by filtering aliquots of 0.2 ml onto Millipore filters and counting the radioactive tracers in the suspension and the filters. B, the effect of Ca2+on the enzyme tryptophan fluorescence was measured as described under "Experimental Procedures." The incubation medium was 20 mM Mops (pH 6.8), 80 mM KCl, 5 mM MgCl,, 1 mM EGTA, and 0.1 mg SR protein/ml. Sequential increments of intrinsic fluorescence were obtained by adding increasing concentrations of Ca2+.The fluorimetric titrations were performed in the absence (0) FIG. 7 . Influence of DES or DES dipropionate on the phosor thepresence of 10 (A), 20 (m), or 30 p~ (0)DES. phoenzyme formed from ATP. A , SR vesicles (0.2 mg/ml) were phosphorylated in an ice bath for 3 s with different concentrations of nate. Fig. 6A indicates that 20 ~ L MDES decreases the Ca2+ [Y-~'P]ATPin a low Caz+reaction mixture containing 20 mM Mops affinity and the maximal binding capacity. Likewise, it can (pH 6.8), 80 mM KCl, 5 mM MgC12, 10 p M A23187, 0.695 mM CaCI2, and 1 mM EGTA. A high Ca'" reaction medium containing 1.99 mM be observed that thesame concentration of DES dipropionate CaClZ was used in the experiments of panel B. The reaction was does not modify any binding parameter. stopped by acid quenching and processed to determine the protein Information of the high affinity Ca2+-binding sites can also and 3zPcontent. 0, control values; A, addition of 20 p M DES; 0, be obtained by fluorimetric titration of the ATPase protein addition of 20 p~ DES dipropionate.

Diethylstilbestrol and Ca2+-transportingATPase DES dipropionate at a concentration 20 pM, the maximal level of phosphorylation was less than expected, although the curves show a similar shape. Thus, DES dipropionate markedly reduced the phosphoenzyme accumulation as a function of the ATP concentration, and DES presented even a more drastic effect on the phosphoenzyme at steady state. When the phosphorylation reaction was performed in a 1 mM free Ca2+medium, the E P formed in the absence of drugs presented the same profile as in the low Ca2+medium (Fig. 7B); however, it was observed that the effect of these compounds on the phosphoenzyme level was considerably smaller. The enzyme phosphorylation by ATP is followed by the vectorial Ca2+ translocationinto thelumen of the vesicles and the subsequent liberation of inorganic phosphate. Therefore, we decided to get information on the rate-limiting step associated to the Ca2+movement by studying the phosphoenzyme decomposition during the enzyme turnover. According to this idea, radioactive E P was built up from [y3'P]ATP at pH6.8, low temperature, and appropriate ionic conditions. A23187 was present to avoid inhibition of the E P decomposition due to the Ca2+accumulated inside the vesicles, and the phosphorylation reaction was maintained for 3 s before dilution with an excess of the same medium, containing nonradioactive ATP (toprevent further formation of 32P-labeledphosphoenzyme). The disappearance of the labeled phosphoenzyme was followed at different time intervals after acid quenching. Fig. 8 shows that DES dipropionate decreases the level of EP accumulated at steady state andaccelerates the rateof decomposition of the phosphorylated species. The kinetic analysis performed in the inset of Fig. 8 indicates that the EPdecay in the presence of20 WM DES dipropionate, showing an apparent rate constant of0.42 s-', is accelerated approximately 1.6-fold with respect to thecontrol experiment. In the case of DES, the E P level accumulated at steady state was less than 1 nmol/mg of protein (see Fig. 7A), making difficult any kinetic characterization of the process. As an alternative, we have successfully monitored the slow transition related to Ca2+release inside the vesicles by measuring the Ca2+dissociation from the phosphorylated enzyme. The experimental approach requires the formation of E P in the presence of ATP and 45Ca2+in leaky vesicles followed by a rapid perfusion of the samples with nonlabeled medium.

The levels of Ca2+bound before the isotopic dilution (zero time in Fig. 9) are in agreement with the Ca2+binding experiments of Fig. 6A. It means that the enzyme retains the maximal capacity of Ca2+binding in the presence of 20 pM DES dipropionate, whereas a similar concentration of DES partially decreases the Ca2+saturation level. The kinetics of 45Ca2+decay as resolved by the rapid filtration technique, which indicates that the apparent rate constant(-10 s-') is increased 1.6-fold when DES dipropionate is present. Moreover, this figure shows that the time course of Ca2+dissociation is practically unaffected by 20 p~ DES. Effect of the Drugs through the Membrane-In an attempt to further characterize the effect of DES and its derivatives on the Ca'+-ATPase protein, we sought membrane probes able to locate in different regions of the lipid matrix and sensitive to our drugs. We found the n-AS molecules to be efficient fluorescent probes for this task. The structures with an inhibitory effect on the Ca2+-ATPase activity, such as DES, dienestrol, or hexestrol promoted a decrease in the fluorescence of n-AS, and the magnitude of this fluorescence change was dependent on the location of the anthroyloxy probe in the stearic acid molecule. This is shown in Fig. lOA, where the decrease of the fluorescence intensity expressed as a percentage was plotted as a function of the hexestrol concentration. The efficiency of the referred effect followed the order 2-AS > 7-AS > 9-AS > 12-AS. Interestingly, DES dipropionate or trans-stilbenehaving an activating action on the enzyme turnover presented the opposite effect, i.e. they caused an increase of the n-AS fluorescence. The extent of this fluorescence change upon increasing the DES dipropionate concentration for different n-AS probes demonstrated (Fig. 10B) that thesequence of the relative effectiveness was 12-AS > 9-AS > 2-AS > 7-AS. DISCUSSION

The ability of different hydrophobic molecules to affect the Ca'+-ATPase activity of SR membranes has already been 10 I

0

. . . . I i!c O0

5

Time

10

(8)

FIG. 8. Time course of phosphoenzyme decomposition at steady state. 32P-labeledphosphoenzyme was formed in an ice-cold Mops bath for 3 s from 25 PM [y3'P]ATP in a medium of20mM (pH 6.8), 80 mM KCl, 5 mM MgC12, 1 mM EGTA, 0.695 mM CaC12, 10 p~ A23187, and 0.2 mg ofprotein/ml in the absence (0)or presence (0)of 20 p M DES dipropionate. Zero time in the figure corresponds t o a 12.5-fold dilution (5 ml) of the samples with ice-cold nonlabeled phosphorylation medium. The inset corresponds to a first-order plot of the experimental data.

11927

4

I

1

0

50

100

Time (ma) FIG. 9. Kinetics of Ca2+ dissociation after enzyme phosphorylation. SR vesicles (0.05 mg/ml) were incubated for 20 s at room temperature in amedium of 20 mM Mops (pH 6.8),80 mM KCl, 5 mM MgCl,, 0.1 mM 45CaC12, 0.1mM EGTA, 10 PM A23187, 1 mM [3H]glucose,and 1 mM ATP. The dissociation process in the millisecond time scale was studied with a rapid filtration apparatus by filtering 2-ml aliquots and perfusing the samples with the phosphorylation medium lacking the radioactive labels. The exchange reaction was carried out in the absence (0)or in the presence of 20 PM DES dipropionate (0)or 20 p M DES (A). Semilogarithmic plots of the experimental data are shown in the inset.

11928

Diethylstilbestrol and Ca2+-transportingA TPase 0

A n

K

Y

0

b

-10

2 t2I b

Y

-20

0

SO

100

DES Dipropionak GY) FIG. 10. Effect of hexestrol ( A )or DES dipropionate ( B )on the fluorescence of n-AS molecules incorporated into the SR membrane. The fluorescence probe (2 p ~ was ) incubated for 1-3 h a t 30 "C in the dark in a medium consisting of 20 mM Mops (pH 6.8), 80 mM KC1, and 10 pg of SR protein/ml. The excitation wavelength was 370 nm and the fluorescence emission was selected with a GG400 cut-off filter. F, represents the initial intensity of the probe incorporated into the membrane and F states for the fluorescence values after the addition of different concentrations of hexestrol or DES dipropionate. The lipophilic drugs were dissolved in ethanol, and a correction for the ethanol addition was made when required. 29-AS (A), or 12-AS (0). AS (O), 7-AS (El),

proved (17-24). A detailed survey reveals that a common characteristic shared by all of these compounds is a high tendency for partitioning into the membrane, giving rise to a membrane concentration-dependent effect. Beyond this fact, the precise lipophilic structure is of paramount importance in determining the activating/inhibitory effect as itis clearly noted in ourstudy. We have selected DES and otherrelated molecules to assess their effect on the functioning of the Ca2+pump. Apart from the clinical relevance of this drug, there is also a basic interest in studying the coupling mechanism between Ca2+transport and ATP hydrolysis and DEScan be a useful tool. Some of the compounds we are dealing with are phenols, and these additives are among the best characterized as interacting molecules with membrane systems. The common pattern response is that the addition oflow concentrations of these compounds promotes a small increase in the enzymatic activity, whereas higher concentrations have a clear inhibitory effect. This behavior is reproducibly obtained in the presence of DES, dienestrol, or hexestrol. Likewise, the half-maximal inhibition of DES is very similar to that observed for H+ATPases isolated from mitochondria (7), chromaffin granules (25), and yeast plasma membrane (26). In order to gain insight into themode of action of DES, we extended our measurements in the presence of other molecules structurally related to DES, namely DES dipropionate and trans-stilbene. Surprisingly, what was observed in the dose-response experiments is that these structures are only

able to activate the Ca2+-dependent ATPase activity. These findings highlight the critical role of the chemical structure in producing modulating effects on the ATPaseprotein through the membrane. A perusal ofFig. 1 indicates that those compounds having hydroxyl groups at para positions on the two benzene rings (i.e. DES, dienestrol, or hexestrol), have the capacity to develop an inhibitory effect. In contrast, DES dipropionate and trans-stilbene having blocked or lacking, respectively, the phenol groups behave only as activators of the hydrolysis rate. Furthermore, the experiments on Ca2+ transport gives a similar pattern to those of the hydrolytic activity confirming the modulating effect of these drugs on the Ca2+-ATPase protein. Functional differences between these two groups of compounds can be unveiled by examining the Ca2+dependence on the ATPase activity in the presence of the tested drugs. The inhibitory molecules decrease the enzyme affinity for Ca2+in the ~ L MCa2+range (Fig. 4A), whereas no effect was observed in the titration range of the low affinity sites. This suggests a selective effect of DES on the high affinity Ca2+ sites. On the other hand, the activating molecules increase the maximal velocity with no change in the affinity of the enzyme for Ca2+in either high and low affinity sites. The membrane protein dependence effect of these drugs suggest a lipid-mediated response (18, 20, 21) or the presence of a very high affinity inhibitor (27). In our experiments, we used a protein concentration of 0.05 mg/ml, which in terms of Ca2+-ATPaseactive sites, represents about0.2 nmol/ml. If we consider that the half-maximal inhibition was obtained with 15 p~ DES, this means that thedrug:enzyme ratio is 75. It is also observed that a change in protein concentration changes the effective concentration range; therefore, several moles of drug must bind to affect the function of one ATPase mole. This suggests a lipid-directed mechanism rather than a specific protein site. Nevertheless, it seems that the interaction of our lipophilic molecules, with the enzyme, occurs through different hydrophobic regions of the proteinembedded into the membrane. This was confirmed by the use of the n-AS fluorescent probes and will be discussed later. Direct measurements of Ca2+ binding indicate that the lipophilic DES is able to interactwith the Ca2+-bindingsites. Since this effect is dependent on the relative concentrations of enzyme and Ca2+,these experiments are not strictly comparable with those of the Ca2+-ATPaseactivity. Nevertheless, we were able to show a clearer effect by decreasing the protein concentration to 0.1 mg/ml andtaking advantage of the fluorescence signal associated to theCa2+-bindingprocess (28, 29). Thus, increasing concentrations of DES cause a decrease in the affinity andthetotal Ca2+ bound tothe protein. Moreover, the presence of the dipropionate derivative that is more lipophilic than DES and expected to be more deeply embedded into the bilayer structure does not interfere with the Ca2+binding sites. The present data suggest a different locus of these drugs into the membrane and support a membranous nature of the Ca2+-bindingsites. This was already suggested by a variety of experimental techniques (30-32). ATP

c E2

ADP

"+ H,o

J

E2P

Pi'

SCHEME I

Diethylstilbestrol and Ca2+-transportingATPase Elucidation of the mechanisms underlying the modulating effects required us to consider the catalytic and transport cycle of the SR ATPase. Scheme I depicts asimplified version of the reaction mechanism where it is assumed that four minimal steps related with the orientation and reactivity of the enzyme during the turnover exist. Under conditionsof PM Ca2+ andlow temperature, the profile of E P accumulation as a function of the ATP concentration was interfered by DES and DES dipropionate. The observation that DES dipropionate diminished the E P level, whereas the ATPase activity was increased, can be explained by an accelerating effect of this compound on the rate-limiting Ca2+ translocation step. This appears to be the case of other activating molecules (33). By contrast, the reduced accumulation of E P and theconcomitant lower enzyme turnover along with the decreased affinity of the enzyme for Ca2+is a clear indication of an inhibitory response of DES before phosphorylation. The differential effect of these compounds on the EPlevels can be eliminated by increasing the Ca2+ concentration atthe mM range, whereas ATP is not able to reestablish the levels of EP. This explains a modulating effect of DES and DES dipropionate on Steps 1 and 3, respectively. Our data on E P decomposition and Ca2+dissociation are of great value in evaluating the effect of these drugs on the enzyme turnover. Namely, the experiments on Ca2+dissociation give direct estimate of the rate-limiting step (Step3). In fact, the apparent rate constant of this step, measured at room temperature, and leaky vesicles coincides with the turnover number measured under similar conditions (34). The chase experiments mentioned above provide similar evidence consistent with a slight activation effect of DES dipropionate and practically no kinetic effect of DES on Step3. Finally, we searched for the modulating domains in the transmembrane (hydrophobic) portion of the enzyme. The location of lipophilic drugs in the membrane structure has already been studied in a number of cases with the aid of fluorescent probes, such as the n-AS. The advantage of the n-AS moleculesis a graded location of the fluorophore at different depths inside the membrane (35-37). These probes, once incorporated into the SR membrane, were sensitive to our lipophilic drugs. Thus, theactivating compounds induced an increase of the n-AS fluorescence, and thiseffect was more important when the 9-anthroyloxy group was located at carbon-12 of the stearic acid molecule. With respect to the inhibitory drugs, it was observed that they decreased the nAS fluorescence intensity, and thiseffect was more important when the fluorescent group was situated at carbon-2 of the stearic acid chain. Under “Results” we only show the effect of hexestrol and DES dipropionate, but all the other structures can also modify the n-AS fluorescence. In any case, the sign of the fluorescence change ( ( F - F,)/F,,) is dependent on the activating or inhibitory actiondisplayed by the drug. The natureof the molecular interaction giving rise to the fluorescence change was not further investigated, although an energy transfer phenomenon can be ruled out since the excitation and emission spectra of the drugs and thefluorescence probes do not overlap. All these findings suggest that the inhibitory drugs have a low capacity of penetration through the phospholipid bilayer and therefore they could be located close to the external part of the lipid matrix inthe region of the Ca2+-bindingsites. The activating drugs will have a preferential location at deeper

11929

positions of the bilayer, and therefore, they will interact with the Ca2+translocation process. REFERENCES 1. Dodds, E. C., Goldberg, L., Lawson, W., and Robinson, R. (1938) Nature 141, 247-248 2. Noller, K. L., and Fish, C. R. (1974) Med. Clin. North Amr. 5 8 , 793-810 3. Ogilvie, M. L., Faltin, E. C., Hauser, E. R., Bray, R. W., and Hoekstra, W. G. (1960) J. Anim. Sci. 19,991-1001 4. Herbst. A. L., Uldfelder, H., andPoskanzer, D. C. (1971) N. End. J. Med. 284,878-881 5. McMartin. K. E.. Kennedv. K. A.. GreensDan, P.. Alam. S. N., Greiner,’P., and Yam, J.”(1978)i.Enuiron. Pathol. Toxicol. 1, 279-313 6. Pedersen, P. L., and Carafoli, E. (1987) Trends Biochem. Sci. 1 2 , 146-150 7. McEnery, M. W., and Pedersen, P. L. (1986) J . Biol. Chem. 2 6 1 , 1745-1752 8. McEnery, M. W., Hullihen, J., and Pedersen, P. L. (1989) J. Biol. Chem. 264,12029-12036 9. Eletr, S., and Inesi, G. (1972) Biochim. Biophys. Acta 282, 174179 10. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275 11. Fabiato, A. (1988) Methods Enzymol. 157, 378-417 12. Schwartzenbach, G., Senn, H., and Anderegg, G. (1957) Helu. Chim. Acta 40, 1886-1900 13. Blinks, J., Wier, W., Hess, P., and Prendergast, F. (1982) Prog. Biophys. Mol. Biol. 4 0 , 1-114 14. Lin, T., and Morales, M. (1977) Anal. Biochem. 77, 10-17 15. Orlowski, S., and Champeil, P. (1991) Biochemistry 3 0 , 352-361 16. Inesi, G., Kurzmack, M., Coan, C., and Lewis, D. (1980) J. Biol. Chem. 255,3025-3031 17. Shoshan, V., and MacLennan, D. H. (1981) J. Biol. Chem. 256, 887-892 18. Ho, M. M., Scales, D. J., and Inesi, G. (1983) Biochim. Biophys. Acta 730, 64-70 19. Hara, H., and Kanazawa, T. (1986) J . Biol. Chem. 261, 1658416590 20. Sokolove, P. M., Albuquerque, E. X., Kauffman, F. C., Spande, T. F., and Daly, J. W. (1986) FEBS Lett. 2 0 3 , 121-126 21. Soler, F., Fernandez-Belda, F., and Gomez-Fernandez, J. C. (1988) Biochem. Biophys. Res. Commun. 1 5 1 , 1093-1098 22. Petretski, J. H., Wolosker, H., and de Meis, L. (1989) J. Biol. Chem. 2 6 4 , 20339-20343 23. Kawashima, T., Hara, H., and Kanazawa, T. (1990) J. Biol. Chem. 265,10993-10999 24. Michelangeli, F., Orlowski, S., Champeil, P., East, J. M., and Lee, A. G. (1990) Biochemistry 29, 3091-3101 25. Gronberg, M., and Flatmark, T. (1988) FEBS Lett. 2 2 9 , 40-44 26. Goffeau, A,, and Slayman, C. (1981) Biochim. Biophys. Acta 639, 197-223 27. Sagara, Y., and Inesi, G . (1991) J. Biol. Chem. 2 6 6 , 13503-13506 28. Dupont, Y.(1976) Biochem. Biophys. Res. Commun. 71,544-550 29. Fernandez-Belda, F., Kurzmack, M., and Inesi, G. (1982) J. Biol. Chem. 259,9687-9698 30. le Maire, M., Lund, S., Viel, A., Champeil, P., and Moller, J. V. (1990) J . Biol. Chem. 2 6 5 , 1111-1123 31. Clarke, D. M., Loo, T. W., Inesi, G., and MacLennan, D. H. (1989) Nature 339, 476-478 32. Sumbilla, C., Cantilina, T., Collins, J. H., Malak, H., Lakowicz, J. R., and Inesi, G. (1991) J. Biol. Chem. 266, 12682-12689 33. Wakabayashi, S., Ogurusu, T., and Shigekawa, M. (1988) J. Biol. Chem. 263,15304-15312 34. Inesi, G., Kurzmack, M., Kosk-Kosicka, D., Lewis, D., Scofano, H., and Guimaraes-Motta, H. (1982) 2. Naturforsch. Teil C Biochem. Biophys. Biol. Virol. 37, 685-691 35. Cadenhead, D. A., Kellner, B. M. J., Jacobson, K., and Papahadjopoulos, K. (1977) Biochemistry 16, 5386-5392 36. Blatt, E., Chatelier, R. C., and Sawyer, W. H. (1984) Photochem. Photobiol. 39, 477-483 37. Blatt, E., and Sawyer, W. H.(1985) Biochim. Biophys. Acta 822, 43-62