Comparison between ATP-supported and GTP-supported phosphate turnover of the calcium-transporting sarcoplasmic reticulum membranes

Quantification and removal of glycogen phosphorylase and other enzymes associated with sarcoplasmic reticulum membrane preparations

The enzymatic characterization of sarcoplasmic reticulum membrane fragments from rabbit skeletal muscle presented in this paper shows that glycogen phosphorylase, as well as other enzymes (e.g., creatine kinase, myokinase, phosphorylase kinase, glycosidase, AMP-deaminase, phosphoglucomutase) are associated with these membrane preparations. Amongst these enzymes, the highest activity associated with sarcoplasmic reticulum membranes is that of glycogen phosphorylase, which is mostly (at least 95%) in its b form (dephosphorylated form), since its activity in sarcoplasmic reticulum membranes is largely dependent upon AMP. A protocol is presented to quantify the amount of phosphorylase bound to sarcoplasmic reticulum membranes from fluorimetric measurements of the content of its coenzyme, pyridoxal 5'-phosphate. The content of phosphorylase ranged from 0.03 to 0.37 mg phosphorylase per mg of membrane protein, in sarcoplasmic reticulum membrane preparations made following several of the protocols most commonly used and also depending upon the length of the starvation period of the animal before killing. We also show that dilution of sarcoplasmic reticulum membranes to 0.1-0.2 mg protein per ml in a buffer containing 50 mM Tes-KOH (pH 7.4), 0.1 M KCl and 0.25 M sucrose removes at least 95% of glycogen phosphorylase from these membrane fragments, as well as other enzymes like myokinase and glycosidase. On these grounds, we suggest to introduce a final dilution step as indicated above in protocols of sarcoplasmic reticulum membrane preparations.

Pressure effects on the binding of vanadate to the sarcoplasmic reticulum calcium-transport enzyme

The effect which hydrostatic pressure exerts on the binding of vanadate to the calcium-transport enzyme was determined. The recent unavailability of radioactive vanadate prevented direct measurements of vanadate binding. The vanadate-free enzyme fraction was instead monitored by phosphorylating it with ATP according to Medda and Hasselbach [Medda, P. & Hasselbach, W. (1983) Eur. J. Biochem. 137, 7-14]. Vanadate binding is reduced with rising pressure at first markedly and subsequently, above 30 MPa, relatively little. The biphasic pressure-binding relationship was analysed by applying a biexponential fitting procedure to the experimental data. The biphasicity of the pressure-binding relationship indicates that the description of vanadate binding requires at least a two-step reaction sequence. The volume increments which predominate at lower pressure values, range from 200-400 ml.mol-1 depending on the composition of the reaction medium containing 5 microM and 20 microM vanadate and no or 15% (by vol.) Me2SO. The binding volumes deduced for the higher pressure range amount to 20-40 ml.mol-1. Vanadate binding is reduced in the presence of 30 microM calcium, and simultaneously both binding volumes are diminished by 100 ml.mol-1 and 20 ml.mol-1 for the low and high pressure values, respectively, as one can expect for mutual interactions between the two ligands of the transport enzyme.

Characterization of the substrate-induced conformational change of N-iodoacetyl-N'-(5-sulfo-1-naphthyl)ethylenediamine-labeled sarcoplasmic reticulum Ca2(+)-ATPase by using different kinds of substrate

Cys-674 of the sarcoplasmic reticulum Ca2(+)-ATPase was labeled with N-iodoacetyl-N'-(5-sulfo-1-naphthyl)ethylenediamine without a loss of the catalytic activity, and changes in the fluorescence intensity upon addition of seven kinds of substrate were followed by the stopped-flow method. The steady-state fluorescence intensity and anisotropy were also determined. When Ca2+ was present, the fluorescence intensity and anisotropy decreased greatly upon addition of any substrate used. The observed affinity for each substrate agreed with the previously observed affinity of the catalytic site. The fluorescence drop induced by the adenine nucleotides, ATP and adenosine 5'-(beta, gamma-methylene)triphosphate (a nonhydrolyzable ATP analog), was much faster than that induced by other substrates. The ATP-induced fluorescence drop preceded phosphoenzyme formation when the ATP concentration was high, but the fluorescence drop coincided with phosphoenzyme formation when it was slowed by reducing ATP concentrations. The fluorescence drop induced by ITP or acetyl phosphate was slow even at high concentrations of the substrate, and it coincided with phosphoenzyme formation. When Ca2+ was absent, the fluorescence intensity and anisotropy decreased only slightly upon addition of any substrate other than the adenine nucleotides. They decreased substantially upon addition of the adenine nucleotides, but the kinetics of this fluorescence drop were quite different from that of the fluorescence drop induced by any substrate in the presence of Ca2+. These results show that the conformational change, which makes the bound label less constrained, is induced by substrate binding to the catalytic site of the Ca2(+)-activated enzyme. This change precedes phosphoenzyme formation in the catalytic cycle and is greatly accelerated by the adenine moiety of the substrate.

The predicted secondary structures of the nucleotide-binding sites of six cation-transporting ATPases lead to a probable tertiary fold

Six cation-dependent transporting ATPases have homologous sequences in the region asigned by chemical labelling to nucleotide binding. Comparison of the most highly conserved segments with other nucleotide-binding domains showed that the sequences were consistent with a mononucleotide-binding fold and enabled a number of likely folding topologies to be limited to two or three alternatives. One of these possible folds was topologically equivalent to adenylate kinase; this was taken as a model in which the significance of conserved amino acids was investigated. In this model conserved amino acids were grouped around a postulated ATP-binding cleft, satisfactorily accounting for their degree of conservation.

Interactions of lipids and proteins: some general principles

Methods to describe the binding of phospholipids to membrane proteins are described. It is shown that it is difficult to obtain estimates of the number of phospholipids bound to the surface of a membrane protein from ESR experiments in which plots of free to bound spin label (y) vs. molar ratio of lipid to protein are extrapolated to y = 0. The relative advantages and disadvantages of ESR and fluorescence methods for measuring relative binding constants of phospholipids to membrane proteins are discussed. The particular problems associated with comparing binding constants of molecules of very different sizes (e.g., fatty acids and cardiolipin) are described and equations are presented to account for these problems. The possible effects of membrane viscosity and thickness on activity of membrane proteins are discussed, but it is concluded that effects of phospholipid structure on activity can only be understood in terms of a reasonably complete kinetic model for the protein.

Characterization of the phosphoenzyme that is involved in the Ca2+ -Ca2+ exchange catalyzed by the Ca2+ -ATPase of sarcoplasmic reticulum vesicles

The rate of Ca2+ efflux was determined with 45Ca2+ -loaded sarcoplasmic reticulum vesicles (mainly with the light fraction of vesicles) at pH 6.5 and 0 degrees C. The efflux depended on external Ca2+, Mg2+, ATP and ADP, but it was not activated by AMP. The results indicate that the efflux is derived from Ca2+ -Ca2+ exchange mediated by the phosphoenzyme (EP) of membrane-bound Ca2+ -ATPase. EP was formed with Ca2+ -loaded vesicles (light fraction) under similar conditions without added ADP. The subsequent addition of EGTA and ADP induced triphasic EP dephosphorylation. Three species of EP (EP1, EP2, and EP3) were distinguished on the basis of this dephosphorylation kinetics, EP1, EP2, and EP3, corresponding to the first, second, and third phases of the dephosphorylation. Dephosphorylation of EP1 and EP2 resulted in stoichiometric ATP formation, while dephosphorylation of EP3 led to stoichiometric Pi liberation. The rate of Ca2+ efflux was compatible with that of EP2 dephosphorylation, whereas it was much lower than the rate of EP1 dephosphorylation and much higher than the rate of EP3 dephosphorylation. The intravesicular Ca2+ concentration dependence of the rate of EP2 dephosphorylation agreed with that of the rate of Ca2+ efflux. The results suggest that isomerization between EP1 and EP2 is the rate-limiting process in the Ca2+ -Ca2+ exchange and that EP3 is not involved in this exchange.

The dependence on internal pH of Ca2+-fluxes across sarcoplasmic reticulum vesicular membranes

The interdependence of the competition between Ca2+ and hydrogen ions for the internally located low-affinity Ca2+ binding sites of sarcoplasmic reticulum vesicles and the pH-dependent splitting rate of phosphoenzyme was investigated. Sarcoplasmic reticulum vesicles were preincubated at a selected pH and passive Ca2+ loading, active Ca2+ uptake at the same pH as well as active Ca2+ uptake at a distinct pH (pH-jump method) were observed. In addition, Cai-Cao exchange in the absence and presence of ADP and ATP-ADP exchange were measured. The overall ATP splitting rate was assayed with leaky vesicles in the presence of varied Ca2+ concentration and four different pH. All experiments were carried out at Ca2+ concentrations sufficient to saturate the externally located activating high-affinity binding sites at all pH and in the absence of affecting concentrations of monovalent cations. Active Ca2+ transport (particularly evident applying the pH-jump method) is facilitated at low intravesicular pH, reflecting the favoured Ca2+ release to the intravesicular space, in contrast to the reverse pH-dependence of passive Ca2+ accumulation and the initial rate of Cai-Cao exchange, both favoured by elevated internal Ca2+ binding capacity. The rates of ATP splitting, the continuing slow rate of Cai-Cao exchange, and the ATP-ADP exchange are optimal at an intermediate proton concentration, reflecting the influence of protons on partial reaction steps occurring later in the reaction cycle and the accelerated exchange of Ca2+ at the internal low-affinity sites as well as the establishment of a new pseudo equilibrium between the possible reaction intermediates. The pool of rapidly exchangeable Ca2+ is enlarged whereas the rate of slow exchange is unaltered or diminished (pH 7.8) by ADP.

Preparation of a highly concentrated, completely monomeric, active sarcoplasmic reticulum Ca2+-ATPase

Sarcoplasmic reticulum vesicles from fast skeletal muscle were partially delipidated with sodium cholate at high ionic strength and sedimented in a discontinuous sucrose gradient. Phospholipid content was reduced from 0.777 mumol/mg protein to 0.242 mumol/mg protein. As judged from gel electrophoresis and high pressure liquid gel chromatography, accessory proteins were removed during centrifugation and the Ca2+-ATPase was obtained in an almost pure form. Addition of myristoylglycerophosphocholine (1 mg/mg protein) reactivates ATPase and dinitrophenylphosphatase activity to the same degree obtained with native vesicles. Using the analytical ultracentrifuge it could be demonstrated that the reactivated Ca2+-ATPase was present exclusively in a monomeric state. These results were obtained at high and low ionic strength and up to a protein concentration of 10 mg/ml. Therefore this preparation should be very useful to investigate differences between oligomeric and monomeric Ca2+-ATPase.

Tryptophan fluorescence of sarcoplasmic reticulum ATPase. A fluorescence quench study

The calcium-dependent change in the tryptophan fluorescence intensity of the sarcoplasmic reticulum Ca2+- and Mg2+-ATPase was investigated using different quenching reagents. It is demonstrated that only those compounds which are bound to the enzyme (i.e., 1-(9,10-dibromomyristoyl)-sn-2-glycerophosphorylcholine and 1-(9,10-dibromostearoyl)-sn-glycero-3-phosphorylcholine) are able to decrease the amplitude of the fluorescence decrement observed after removal of calcium ions. From the position of the bromine atom within the lysophosphatidylcholines, it is concluded that the tryptophan residues involved are located in the hydrophobic part of the ATPase molecule and are in contact with the hydrocarbon chains of the phospholipids.

Undirectional calcium and nucleotide fluxes in cardiac sarcoplasmic reticulum. II. Experimental results

Unidirectional calcium influx and efflux were evaluated in cardiac sarcoplasmic reticulum (SR) by 45Ca-40Ca exchange at steady state calcium uptake in the absence of calcium precipitating anions. Calcium efflux was partitioned into a pump-mediated efflux and a parallel passive efflux by separately measuring passive efflux referable to the steady state. Unidirectional and net ATP-ADP fluxes were measured using [3H]-ATP----ADP and [3H]-ADP----ATP exchanges. Methods are presented that take into account changing specific activities and sizes of the nucleotide pools during the measurement of nucleotide fluxes. The contribution of competent and incompetent vesicles to the unidirectional and net nucleotide fluxes was evaluated from the specific activity of these fluxes in incompetent vesicles and from the fraction of vesicles that were incompetent. The results indicate that, in cardiac SR, unidirectional calcium fluxes are larger than the unidirectional nucleotide fluxes contributed by competent vesicles. Because the net ATPase rate of competent vesicles is similar to the parallel passive efflux, it appears that cardiac SR Ca-ATPase tightly couples ATP hydrolysis to calcium transport even at static head, with a coupling ratio near 1.0.

Unidirectional calcium and nucleotide fluxes in sarcoplasmic reticulum. I. Interpretation of flux ratios for different reaction schemes

The relation between unidirectional calcium and nucleotide fluxes was examined for different ATPase reaction schemes of the sarcoplasmic reticulum. The schemes considered differed in the order of sorption and desorption of calcium, ATP, and ADP. The results suggest that the theoretical relation between calcium and nucleotide fluxes depends on the reaction scheme and that experimental measurements can distinguish among them. The results obtained are generally valid and do not depend on assumptions of equilibrium or pseudoequilibrium between intermediate states of the pump.

Inactivation of the calcium-transport ATPase in the sarcoplasmic reticulum by the combined effect of lasolocid and Triton X-100

The calcium-transport ATPase of the sarcoplasmic reticulum membranes is irreversibly inactivated by the combined action of Lasolocid and Triton X-100 at concentrations which separately do not interfere with the enzyme's activity. In the presence of Lasolocid the enzyme is most susceptible to inactivation when the Triton X-100 concentration just exceeds its critical micellar concentration, approximately, 0.2 mg X ml-1. Lasolocid becomes effective at a concentration of 10 microM and produces rapid inactivation at 100 microM. Phosphoprotein formation is less affected than phosphate liberation. The influence of the ATPase protein on the fluorescence intensity of Lasolocid passes a distinct maximum at the most effective Triton X-100 concentration of 0.2 mg X ml-1.

Determinants of calcium loading at steady state in sarcoplasmic reticulum

The determinants of steady-state calcium loading by sarcoplasmic reticulum vesicles were evaluated by measuring the contribution of different pathways of calcium flux to the total calcium flux at steady state. The diffusional passive pathway was least significant at all calcium loads studied. Diffusional passive calcium flux was evaluated by a number of methods which gave comparable results and support its designation as passive and diffusional. These methods included (a) flux measurements with the simple pump-leak system which pertains when acetyl phosphate is used to load the vesicles; (b) flux measurements made after quenching the pump with EGTA; (c) flux measurements made after quenching the pump with glucose plus hexokinase; and (d) evaluation of the effect of pump activity on the efflux of mannitol. The calcium efflux not accounted for by the diffusional pathway was assigned to non-diffusional pathways. Efflux through the non-diffusional pathways required ATP, ADP and extravesicular Ca2+. The ADP-dependent, phosphoenzyme-independent pathway described by Beirao and DeMeis (Biochim. Biophys. Acta (1976) 433, 520-530) was not significantly involved in efflux. We propose that the level of calcium loading achieved at steady state is determined by the levels of the intermediates of the calcium pump which are established at this pseudo-equilibrium condition, these levels being determined by the concentrations of intravesicular and extravesicular calcium ([Ca2+]i and [Ca2+]), ATP and ADP. The different levels of calcium loading achieved by skeletal and cardiac sarcoplasmic reticulum are attributed to different nucleotide and calcium kinetics in these two types of sarcoplasmic reticulum and possibly to different intravesicular volumes. Differences in diffusional permeability are not responsible for differences in calcium loading.

Excimer formation of ATPase from sarcoplasmic reticulum labeled with N-(3-pyrene)maleinimide

Sarcoplasmic reticulum ATPase from fast skeletal muscle was labeled in native vesicles with N-(3-pyrene)maleinimide. At labeling ratios larger than 1 mol pyrenemaleinimide/2.5 mol ATPase significant amounts of excimers are detected. Excimer concentration decreases at low, non-solubilizing amounts of detergents (0.2 mg X mg protein-1) and completely disappears after solubilization of the membranes. These results exclude that excimers are formed due to 'double-labeling' of one ATPase molecule. It is concluded that the ATPase exists as an oligomer within the membrane of native vesicles.

ATP-ADP exchange reaction by fragmented sarcoplasmic reticulum from bullfrog skeletal muscle

The ATP-ADP exchange reaction and its related partial reactions of fragmented sarcoplasmic reticulum from bullfrog skeletal muscle (frog FSR) were investigated and compared with those of rabbit FSR in order to understand the characteristics of calcium-activated ATPase (Ca2+-ATPase) of frog FSR. MgATP and magnesium-free ADP are substrates for the forward and backward reaction of the ATPase activity, respectively, which is consistent with the conclusion obtained with rabbit FSR. The ATP-ADP exchange rate of frog FSR increased sharply with an increase in Ca2+ concentration up to 3 microM, and then decreased as Ca2+ concentration increased from 3 microM to 100 microM, where the level of EP continued to increase. The exchange rate of frog FSR had a value similar to the overall ATPase activity at steady state. These results contrast with observations using rabbit FSR. The exchange rate of rabbit FSR, which is 10-30 times as high as the overall ATPase activity, reached a plateau at 1 microM Ca2+, and the decrease in the exchange rate with the increase in Ca2+ concentration was not observed until the concentration was greater than 30 microM, where the plateau of the ATPase activity was maintained. These results were discussed in reference to a possible ordered reaction sequence of ATP followed by calcium in the Ca2+-ATPase reaction. It is suggested that k-5/k-6 for rabbit FSR at steady state should be larger than that for frog FSR by a factor of about 10 in the following reaction sequence. (Formula see text).

Formation of magnesium-phosphoenzyme and magnesium-calcium-phosphoenzyme in the phosphorylation of adenosine triphosphatase by orthophosphate in sarcoplasmic reticulum. Models of a reaction sequence

The aim of the present study was to test simple reaction sequences which describe calcium-independent plus calcium-dependent phosphorylation of sarcoplasmic reticulum transport. ATPase by orthophosphate including the function of magnesium in phosphoenzyme formation. The reaction schemes considered were based on the reaction sequence for calcium-independent phosphorylation proposed previously; namely that the transport enzyme (E) forms a ternary complex (Mg . E . Pi), by random binding of free magnesium and free orthophosphate, which is in equilibrium with the magnesium-phosphoenzyme (Mg . E-P). Phosphorylation, performed at pH 7.0 20 degrees C and a constant free orthophosphate concentration using sarcoplasmic reticulum vesicles either unloaded or loaded passively with calcium in the presence of 5 mM or 40 mM CaCl2, resulted in a gradual decrease in the apparent magnesium half-saturation constant and an increase in maximum phosphoprotein formation with increasing calcium loads. When phosphorylation of sarcoplasmic reticulum vesicles preloaded in the presence of 5 mM CaCl2 was performed at a constant free magnesium concentration, a decrease in the apparent orthophosphate half-saturation constant and an increase in maximum phosphoprotein formation was observed as compared with vesicles from which calcium inside has been removed by ionophore X-537A plus EGTA treatment; however, both parameters remained unchanged by increasing free magnesium from 20 mM to 30 mM. When phosphorylation of sarcoplasmic reticulum vesicles passively loaded with calcium in the presence of 40 mM CaCl2, at which the saturation of the low-affinity calcium binding sites of the ATPase is presumably near maximum, was performed at increasing concentrations of free orthophosphate, there was a parallel shift of phosphoprotein formation as a function of free magnesium and vice versa, with no change in the maximum phosphoenzyme formation. Comparison of the experimental data with the pattern of phosphoprotein formation predicted from model equations for various theoretical possible reaction sequences suggests that phosphoenzyme formation from orthophosphate possesses the following features. Firstly, calcium present at the inside of the sarcoplasmic reticulum membrane binds to the free enzyme and in sequential order to E . Mg . Pi or Mg . E-P or to both, but neither to E. Mg nor to E . Pi. Secondly, calcium-independent and calcium-dependent phosphoproteins are magnesium-phosphoenzymes. Calcium-dependent phosphoenzyme is a magnesium-calcium-enzyme phosphate complex with 1 magnesium, 2 calciums and 1 orthophosphate (the last covalently) bound to the enzyme [Mg . E-P . (Cai)2], and not a 'calcium-phosphoprotein' without bound magnesium.

Ca2+ uptake, Ca2+-ATPase activity, phosphoprotein formation and phosphate turnover in a microsomal fraction of smooth muscle

Vesicles capable of phosphate-stimulated calcium uptake were isolated from the microsomal fraction of the smooth muscle of the pig stomach according to a previously described procedure which consists in increasing the density of the vesicles by loading them with calcium phosphate and isolating them by centrifugation [Raeymaekers, L., Agostini, B., and Hasselbach, W. (1981) Histochemistry, 70, 139--150]. These vesicles, which contain calcium phosphate deposits, are able to accumulate an additional amount of calcium. This calcium uptake is accompanied by calcium-stimulated ATPase activity and by the formation of an acid-stable phosphoprotein. The acid-denatured phosphoprotein is dephosphorylated by hydroxylamine, which indicates that an acylphosphate is formed. This phosphoprotein probably represents a phosphorylated transport intermediate similar to that seen with the Ca2+-ATPase of sarcoplasmic reticulum of skeletal muscle. As with the Ca2+-ATPase of sarcoplasmic reticulum vesicles, this vesicular fraction catalyses an exchange between inorganic phosphate and the gamma-phosphate of ATP (ATP-Pi exchange) which is dependent on the presence of intravesicular calcium, and an exchange of phosphate between ATP and ADP (ATP-ADP exchange). The results further indicate that the turnover rate of the calcium pump, calculated from the ratio of calcium-stimulated ATPase activity to the steady-state level of phosphoprotein, is similar to that of Ca2+-ATPase of sarcoplasmic reticulum of skeletal muscle.

Preparative isolation of Apo(Ca2+-ATPase) from sarcoplasmic reticulum and the reactivation by lysophosphatidylcholine of Ca2+-dependent ATP hydrolysis and partial-reaction steps of the enzyme

1. A preparative method for the isolation of the lipid free apoprotein, the Ca2+-ATPase of sarcoplasmic reticulum, from the partially purified lipoprotein, Ca2+-ATPase vesicles, is presented. 2. By enzymatic hydrolysis of the phospholipids and removal of the splitting products and endogenous neutral lipids, the apoprotein was consistently delipidated to 0.02 mumol Pi/mg protein. 3. Reactivation of the splitting of ATP and the pseudo substrate, dinitrophenyl phosphate, was demonstrated with a variety of lipids and detergents. 4. A total reactivation of ATP splitting was achieved after a mild ultrasonication of the apoprotein with myristoylglycerophosphocholine which resulted in solubilization of the enzyme as an optically clear solution. 5. The stable resolubilized enzyme could be stored for several weeks maintaining full enzymatic activity. Gel chromatography suggested that under the assay conditions, the monomeric form of the enzyme predominated. 6. In comparison with the native enzyme, the resolubilized enzyme showed differences in the temperature dependence of the activation of ATP hydrolysis and a reduced apparent affinity for MgATP. 7. The phosphate-transferring activities of the resolubilized enzyme were only partially reactivated in the forward direction, and none of the reverse partial-reaction steps of the enzyme could be demonstrated.

Variable Ca2+ transport:phosphoprotein ratios in the early part of the GTP-driven calcium-transport reaction of the sarcoplasmic reticulum

Initial Ca2+ transport and phosphoprotein formation of the sarcoplasmic reticulum membrane with GTP were investigated in a comparative study. While saturation of the high-affinity sites for Ca2+ binding and transporting as well as for GTP binding on the external surface of the membrane resulted in Ca2+ transport and phosphoprotein formation in a molar ratio of 2, the variation of the concentrations of the two reactants yielded ratios between 1.7 and 5.7. The ratios varied with a similar dependence on the concentrations of Ca2+ and GTP, except at 500 microM Ca2+, if the reaction was started by Ca2+ instead of GTP but the overall rates decreased. 1 mM DL-propranolol in the preincubation medium selectively inhibited Ca2+ transport but had no effect on initial phosphoprotein formation. These observations indicate that:L (a) phosphorylation of one enzyme molecule induces Ca2+ transport by a variable but limited number of neighbouring molecules, (b) not all Ca2+ bound is essential for phosphorylation but can be transported in parallel, (c) Ca2+ bound to low-affinity sites occupied at 500 microM Ca2+ in the reaction medium is also transported initially, (d) the accessibility of the high-affinity Ca2+ binding sites for DL-propranolol differs, (e) DL-propranolol interacts with Ca2+ binding and transporting sites only in that conformation of the enzyme that can be phosphorylated by the nucleotide.