Changes in Ca2+ Affinity Related to Conformational Transitions in the

0 downloads 0 Views 1MB Size Report
Jan 10, 2016 - chromatography on a Sephadex G-50 column for detergent-solubilized as well as .... The sample (1 ml of the same medium containing.
Val. 260,No. 1, Issue of January 10, p : 371-360,1985

THEJOURNAL OF BIOLOGICAL CHEMISTRY

nnted In U.S.A.

0 1985 by The American Society of Biological Chemists, Inc.

Changes in Ca2+Affinity Related to Conformational Transitions in the Phosphorylated State of Soluble Monomeric Ca2+-ATPase from Sarcoplasmic Reticulum* (Received for publication, July 11, 1984)

Jens P. Andersen, Kirsten Lassen, and JesperV. M ~ l l e r From the Instituteof Medical Biochemistry, University of Aarhw, 8000 Aarhus C, Denmark

Changes in Ca2+binding after phosphorylation of membranous or detergent-solubilized preparations of sarcoplasmic reticulum Ca2+-ATPasewith ATP were followed spectrophotometrically by the use of murexide. Distinct Caz+ release from the two high-affinity translocation sites wasobserved, particularly at alkaline pH and at lowCa2+/Mg2+ concentration ratios. Phosphorylation also induced additional binding of Ca2+ at a third site in competition with M 8 + . Ca2+ releasewas increased after solubilization of Ca2+ATPase in predominantly monomeric form with the nonionic detergent octaethyleneglycol monododecyl ether. At 0 OC, chemical-quench studies with[32P]ATP indicated that release of Ca2+ is correlated with the level of ADP-insensitive phosphoenzyme (2 mol ofCa2+ released per mol of EzPformed), both for membranous and detergent solubilized Ca2+-ATPase.Ca2+release was also found to be accompanied bychanges in intrinsic fluorescence. Analysis of the data at20 "C, pH 8.0, showed that binding of Ca2+to transport sites on E2P occurs with a half-saturation constantof 0.7 mM and a Hill coefficient of 1.8.This isconsistent with a drastic decrease in Ca2*affinity following conversion of ADPsensitive EIP to ADP-insensitive E2P. The similarity between membranous and detergent-solubilized Ca2+ATPase supports the view that not more than a single Ca2+-ATPasepolypeptide chain is requiredto complete the conformational transitions which are thebasis for active transportof Ca".

The sarcoplasmic reticulumCa2+-ATPaseis capable of transporting Ca2+ against a large concentration gradient at the expense of free energy released by hydrolysis of ATP or other energy-rich phosphates (1, 2). The structural basisfor active ion accumulation isa fundamental, but yet unresolved problem (3-5). Various models have been proposed in which the Ca2+pump is oligomeric to accountfor evidence of kinetic heterogeneity of the Ca2+-ATPasepolypeptides (6-9), disappearance of metal sensitivity after solubilization with detergent (lo), or a target size for radiation inactivation of 210,000250,000 daltons (11).On the other hand,a number of studies on detergent-solubilized Ca2+-ATPasehaveindicatedthat monomeric Ca2+-ATPase is capable of hydrolyzing ATP in basically the sameway as membranous Ca2+-ATPase(12-17) and of undergoing conformational transitions related ligand to

binding (16, 18). An unanswered aspect is whether in these noncompartmentalized systems ATPhydrolysis is coupled to translocation of Ca2+. This requires transfer of free energy from bound phosphate Ca", to presumably mediated through a protein-conformational change (19). After translocation of Ca2+,the Ca2+-ATPase is expected to have a decreased affinity for Ca2+(5). In agreement with this view, it hasbeen possible under appropriate conditions to demonstrate release of Ca2+ afterphosphorylation of membranousCa2+-ATPase with ATP (20-23). Studies on detergent-solubilized Ca"-ATPase haveled to conflictingresults: Deanand Gray(24) were unable to demonstrateCa2+ releaseafter disintegrationof the membrane structure with the nonionic detergent C12Es,'while Takisawa and Makinose (25, 26) observed Ca2+release after solubilization with deoxycholate at high Mg2+ concentrations. The purpose of this investigation has been 2-fold 1) To study phosphorylation-induced Ca2+ release after solubilization of Ca2+-ATPase inmonomeric form with detergent. This was done in order to gain insight into the structural basis (monomeric uersus oligomeric) for Ca2+ translocation. 2) To study ina comprehensive way conditions which optimize Ca2+ release for membranousas well as detergent-solubilized Ca2+ATPase. Ca2+translocation is supposed to occur in relation to theconversion of ADP-sensitive phosphoenzyme ( EIP) to ADP-insensitive phosphoenzyme (E,P) (1).It is, therefore, important toknow whether any observed Ca2+ releasecan be related quantitatively to the formation of E2P. This aspect was studied by chemical-quench methods at 0 "C. In addition, the use of intrinsic fluorescence allowed us to estimate the fraction of phosphoenzyme present as E2P at 20 'C, where the kinetic properties of phosphorylated Ca2+-ATPase are less well established. EXPERIMENTALPROCEDURES

Preparation and Characterization of Ca2+-ATPase-Sarcoplasmic reticulum vesicles were prepared from rabbit skeletal muscle as described by de Meis and Hasselbach (27).Purified, leaky Ca2+-ATPase vesicles were obtained from this preparationby extraction with a low concentration of deoxycholate according to Method 2 of Meissner et al. (28). The maximum amount of phosphoprotein formed from ATP was 4.24 nmol/mg f 0.39 ( n = 8 ) in eight different preparations and was not changed significantly by solubilization with C12EB(Nikko Chemicals, Tokyo) at a weight ratio of lO:l, except after exposure to acid and alkalinemedia outside the pH range of 6.5-8.5. High-affinity Ca2+binding in absence of ATP, as measured by Millipore filtration (see below), was 8-9 nmol/mg. The same value was obtained by gel chromatography ona Sephadex G-50 columnfor detergent-solubilized *This investigation has been supported by the Danish Medical as well as membranous Ca2+-ATPase. Hence,we adopted a value of Research Council, the Fogh-Nielsen Foundation, and the Foundation of King Christian X. The costs of publication of this article were The abbreviations used are: CI2E8, octaethyleneglycol monododefrayed in part by the payment of page charges. This article must decyl ether; Tes, N-{tris[hydroxymethyl]methyl-2-amino)ethanetherefore be hereby marked "aduertisement" in accordance with 18 sulfonic acid EP, phosphorylatedCa2+-ATPase;EGTA,ethylene glycol bis(P-aminoethyl ether)-N,N,N',N'-tetraaceticacid. U.S.C. Section 1734 solely to indicate this fact.

371

372

Changes in Ca2+Affinity of

Soluble Monomeric Ca2+-ATPase

8.5 nmol of Ca2+/mgfor the concentration of the Ca'+-translocation dephosphorylation was followed, after various time intervals, by sites. This binding capacity is lower than that estimated for a pure quenching with 2.0 ml of ice-cold 8% (w/v) perchloric acid and 10 preparation, suggesting the presence of inactive Ca2+-ATPase (29, mM pyrophosphate (34). Precipitation was allowed to occur for 30 30). In accordance with this view, we have found, by gel chromatog- min a t 0 "C and then precipitate was washed three times with 5 ml raphy ona Bio-GelA-1.5m column, that in the presence of a relatively of 0.1% trichloroaceticacid, 10 mM pyrophosphateand 10 mM low concentration of C12Es (0.5 mg/ml in the eluant) approximately KHZPO, by centrifugation for 30 min a t 20,000 X g. Finally, the one-third of the Ca2+-ATPase preparation solubilized a t 2 mg of protein wasresuspended in 5% sodium dodecyl sulfate and 0.2 M CI2E8/mgof protein separates from active enzyme in an inactive and NaOHandheated a t 55"C for 20 min. Aliquots were taken for aggregated form (31). measurement of radioactivity and protein content. Analytical Ultracentrifugation-The aggregational state of solubiA similar procedure was followed when phosphoprotein was aslized Ca2+-ATPasewas examined by sedimentation velocity as presayed directlywithout dephosphorylation. In thiscase, acid quenching viously described (15, 16) ina Beckman Model E analytical ultracen- was performed 30 s (0 "C) or 5 s (20 "C) after addition of ATP. For trifuge equippedwitha photoelectric scanner and a temperaturedetermination of the maximum amount of phosphoprotein, phosphocontrolunit.Estimation of thefraction of soluble Ca2+-ATPase, rylation was performed with 100 yM [ Y - ~ ~ P ] A at T P0.5 mM Ca2+, 10 which is monomeric, is based on evaluation of scans showing protein mM M%+, 50 mM Tes/Tris (pH 8.0), and 0.1 M KC1 a t 0 "C. concentration (monitored as absorbance a t 280 nm) as a function of Ca2+-binding Measurements-Absolute levels of Ca2+binding, in radial distance, and evaluation of the first derivative (concentration the presence or absence of ATP, were obtained by a Millipore doublegradient) calculated from these scans. The monomeric fraction, hav- filter technique. To 10 ml of medium (50 mM maleate/Tris, 0.1 M ing a sedimentation coefficient ( s ~ ~of, 4.5-5.0 ~ ) S was well separated KCI, 50 p~ "Ca2+ adjusted to the pH and M$+ concentration shown fromaheterogenous oligomeric component, whichcomprised less in the figures) was added 0.4 mg of ATPase, and, after 20 s, the than 30% of the protein in the standard conditions of the Ca2+-release mixture was filtered through two 0.45 FM Millipore filters at 20 "C. experiments (6.0 mg of C12E8/ml,1.0 mg of protein/ml). Variation in In experiments including ATP, the concentration of nucleotide was Ca2+,M%+, K+, and H+ concentrations as in the Ca2+-release exper- 30 p~ and the medium, in addition, contained 1 mM phosphoenoliments did not change the amountof monomeric Ca2+-ATPase pres- pyruvate and40 IU of pyruvate kinase/ml. Radioactivity onthe filters ent. Similarly, a decrease in temperature from 20 to 2 "C was without was measured by liquid scintillation counting, and membrane-bound influence. In the latter case, the sedimentation coefficient of the Ca2+was calculated after subtractionof the radioactivity on thelower major component decreased, but thevalue corrected for viscosityand filter fromthat on the upper filter.For C12E8solubilized CaZ+-ATPase, density effects ( s ~ , , . ~was ) 4.5 S,consistent with a monomer (15). a Sephadex G-50 column (1 X 25 cm) was used for measurement of ATP-induced Release and Bindingof Ca2+-Changes in Ca2+bind- Ca2+binding. The column was equilibrated a t 20 "C with 6.0 mg of ing during phosphorylation with ATP were measured by recording C12E8/ml,50 mM Tes/Tris (pH 8.0), 0.1M KC1, 50 p M "Ca2+, with or changes in the absorptionof murexide (Merck) a t 544 nm, relative to without 30 p~ ATP. The sample (1ml of the samemedium containing 495 nm, on an Aminco DW-2 spectrophotometer equipped with an 3.0 mg of Ca2+-ATPase and 24 mg of C12Es)was chromatographed efficiently stirredcuvette (32). Thestandard composition of the during less than 1 min and 1-ml fractions analyzed for protein and medium used for these measurements was 50 mM Tes/Tris (pH 8.0), radioactivity. For estimation of the capacity for high-affinity Ca2+ 0.1 M KCI, 50 p M ca2+, 10 mM M$+, 0.1 mM murexide, 1.0 mgof binding by a similar procedure, 10 mM Mg2+ and no ATPwas present Ca2+-ATPase protein/mla t 20 "C. For solubilization of Ca2+-ATPase, in thebuffers. the mediumalso contained 6.0 mg ofC12E8/ml. Thesestandard Other Methods-Ca2+-ATPase activity was measured at 20 "C by conditions were varied as described in the legends to the figures and a NADH-coupled assay (15). The medium contained 50 mM Tes/Tris tables. The Ca2+ concentration was adjusted by taking into consid(pH 8.0), 0.1 M KCI, 50 yM Ca2+,M%+ as indicated in the figure, 30 eration the amountof contaminant Ca2+ present in the Ca2+-ATPasep~ ATP, 1 mM phosphoenolpyruvate, 0.2 mg of pyruvatekinase preparation (20-30 nmol/mg)andin buffer solutions (-5 p M ) as (Sigma, 500 IU/mg)/ml, 0.15 mM NADH, 0.06 mg of lactate dehydrodetermined by atomic-absorption spectrometry. Ca2+release was in- genase (Boehringer Mannheim, 550 IU/mg)/ml and 0.01-0.05 mg of itiated by addition of ATP in a volume which was less than 0.1% of Ca2+-ATPase/ml, with or without 0.5 mg C12E8/ml.Protein concenthe total sample volume. Changes in absorption were converted to trations were determined by the method of Lowry et al. (35) standCa2+concentration by relatingthe response tothatobtained by ardized as previously described (16). ATP concentration was deteraddition of solutions ofCaC12. At high Ca2+/M%+-concentration mined by the luciferase assay (32). ratios it was necessary to correct the data for Ca2+binding to free ATP, as measuredby the spectrophotometricresponse. RESULTS Fluorescence Measurements-Tryptophan fluorescence was measured in a Perkin-Elmer M P F 44A spectrofluorimeter as previously Changes in Caz+ Affinity during A T P Hydrolysis-Fig. 1 described (16).The conditions were similar to those of Ca2+-release shows changes inmurexide absorption induced by addition of experiments, except that murexidewas absent and thatlower protein and detergent concentrations (0.2 and 0.5 mg/ml, respectively) were ATP to purified, membranous Ca2+-ATPase (leaky vesicles) of the present experiments employed. The mixing time, as well as the time constantfor registra- a t 20 "C under standard conditions (50 ptM Ca2+,10 mM Mg+,0.1 M KCl, pH 8.0). This results in tion of fluorescence changes, was much longer than the values pertaining to the Ca2+-release experiments. Therefore,slowness the with a temporary Ca2+releasefrom the Ca2'-ATPase which is which the steady-state fluorescence level was reached (Fig. 7) does associated with phosphorylationof the enzyme. The phosphonot reflect a true difference between the rate of fluorescence change rylation level during Ca2+ releaseis 4-4.5 nmol/mg of protein, and Ca2+release. Stopped-flow experiments' have indicated that the of active enzyme present in the fluorescence changes are actuallycompleted within 1 s at 20 "C, as is corresponding to the amount preparation. After some seconds, murexide absorption gradthe case for Ca2+ release. In the figures, the fluorescencechanges ( A F ) have been expressed as percentages of the total fluorescence ually returns to the samelevel as before addition of ATP. At ( F ) observed before addition of ATP. All fluorescence changes in- this time, all ATP is hydrolyzed (