Circular dichroism and fluorescence studies on interaction of calmodulin (CaM) with purified (Ca2(+)-Mg2+)ATPase of erythrocyte ghosts

It was found, using circular dichroism spectroscopy, that CaM, in the presence of Ca2+, decreases the alpha-helix content of (Ca2(+)-Mg2+)ATPase of porcine erythrocytes from 66% to 55%. In the absence of Ca2+ the enzyme showed 46% of alpha-helix. Moreover, quenching of the ATPase intrinsic fluorescence by acrylamide indicated that, depending on the enzyme conformational status, the accessibility of its tryptophan residues is influenced by direct interaction with CaM at micromolar Ca2+ concentration. This was also confirmed by the observation that fluorescence energy transfer occurred from tryptophan residues of (Ca2(+)-Mg2+)ATPase to dansylated CaM. The presented results may indicate that binding of CaM gives rise to a novel conformational state of the enzyme, distinct from E1 and E2 forms of the Ca2+ pump.     

Troponin C/calmodulin chimeras as erythrocyte plasma membrane Ca2+-ATPase activators

Calmodulin (CaM) and troponin C (TnC) are EF-hand proteins that play fundamentally different roles in animal physiology. TnC has a very low affinity for the plasma membrane Ca2+-ATPase and is a poor substitute for CaM in increasing the enzyme's affinity for Ca2+ and the rate of ATP hydrolysis. We use a series of recombinant TnC (rTnC)/CaM chimeras to clarify the importance of the CaM carboxyl-terminal domain in the activation of the plasma membrane Ca2+-ATPase. The rTnC/CaM chimera, in which the carboxyl-terminal domain of TnC is replaced by that of CaM, has the same ability as CaM to bind and transmit the signal to Ca2+ sites on the enzyme. There is no further functional gain when the amino-terminal domain is modified to make the rTnC/CaM chimera more CaM-like. To identify which regions of the carboxyl-terminal domain of CaM are responsible for these effects, we constructed the chimeras rTnC/3CaM and rTnC/4CaM, where only one-half of the C-terminal domain of CaM (residues 85-112 or residues 113-148) replaces the corresponding region in rTnC. Neither rTnC/3CaM nor rTnC/4CaM can mimic CaM in its affinity for the enzyme. Nevertheless, with respect to the signal transduction process, rTnC/4CaM, but not rTnC/3CaM, shows the same behaviour as CaM. We conclude that the whole C-terminal domain is required for binding to the enzyme while Ca2+-binding site 4 of CaM bears all the requirements to increase Ca2+ binding at PMCA sites. Such mechanism of binding and activation is distinct from that proposed for most other CaM targets. Furthermore, we suggest that Ala128 and Met124 from CaM site 4 may play a crucial role in discriminating CaM from TnC.     

Purification of the (Ca2+-Mg2+)-ATPase from human erythrocyte membranes using a calmodulin affinity column.

The (Ca2+-Mg2+)-ATPase from human erythrocyte membranes has been solubilized in Triton X-100 and purified on a calmodulin affinity chromatography column in the presence of phosphatidylserine, to limit the inactivation of the enzyme. The enzyme was purified at least 150 times when compared with the original ghosts and showed a specific activity of 3.8 mumol.mg-1.min-1. In sodium dodecyl sulfate-polyacrylamide gels, a single major band was visible at a position corresponding to a molecular weight of about 125,000; a minor band (11% of the total protein) was present at a position corresponding to Mr = 205,000. Upon incubation of the purified preparation with [32P]ATP, both bands were phosphorylated in proportion to their mass, suggesting that both were active forms of purified ATPase.     

Activation of an islet cell plasma membrane (Ca2+ + Mg2+)-ATPase by calmodulin and Ca-EGTA

Islet cell plasma membranes contain a calcium-stimulated and magnesium-dependent ATPase (Ca2+ + Mg2+)-ATPase) which requires calmodulin for maximum enzyme activity (Kotagal, N., Patke, C., Landt, M., McDonald, J., Colca, J., Lacy, P., and McDaniel, M. (1982) FEBS Lett. 137, 249-252). Investigations indicated that exogenously added calmodulin increases the velocity and decreases the Km for Ca2+ of the high affinity (Ca2+ + Mg2+)-ATPase. These studies routinely employed the chelator ethylene glycol bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) to maintain Ca2+ concentrations in the submicromolar range. During the course of these investigations, it was found unexpectedly that increasing the concentrations of EGTA (0.1-4 mM) and total calcium in the media, while maintaining constant free Ca2+ levels, increased the velocity of the high affinity (Ca2+ + Mg2+)-ATPase. The free calcium concentrations under these conditions were verified by a calcium-sensitive electrode. The (Ca2+ + Mg2+)-ATPase maximally activated by 2-4 mM EGTA was not further stimulated by calmodulin, whereas camodulin stimulation increased as the concentration of EGTA in the media was decreased. A similar enhancement by Ca-EGTA was observed on active calcium transport by the plasma membrane-enriched fraction. Moreover, Ca-EGTA had a negligible effect on both active calcium transport as well as Ca2+-stimulated ATPase activity by the islet cell endoplasmic reticulum, processes which are not stimulated by calmodulin. The results indicate that stimulation by Ca-EGTA may be used to differentiate calcium transport systems by these subcellular organelles. Furthermore, the concentration of EGTA routinely employed to maintain free Ca2+ levels may itself obscure effects of calmodulin and other physiological agents on calcium-dependent activities.     

Inhibition of membrane erythrocyte (Ca2+ + Mg2+)-ATPase by hemin

Red blood cell lysis is a common symptom following severe or prolonged oxidative stress. Oxidative processes occur commonly in sickle cells, probably mediated through denatured hemoglobin and the accumulation of ferric hemes in the membranes. Calmodulin-stimulated (Ca2+ + Mg2+)-ATPase from sickle red cell membranes is partially inactivated (Leclerc et al. (1987) Biochim. Biophys. Acta 897, 33-40). In this study (Ca2+ + Mg2+)-ATPase activity from normal adult erythrocyte membranes was measured in the presence of hemin. We report a time- and concentration-dependent inhibition of the activity of the enzyme by hemin due to a decrease in the maximum velocity. Only a mild inhibitory effect was observed in the presence of iron-free protoporphyrin IX, indicating the catalytic influence of the iron. Experiments carried out with hemin (ferric iron) liganded with imidazole or with reduced protoheme (ferrous iron) liganded with carbon monoxide, demonstrated that the inhibition requires that hemin be capable of binding additional ligands. The inhibition was not influenced by the absence of oxygen but was prevented by addition of bovine serum albumin. Addition of butylated hydroxytoluene, a protective agent of lipid peroxidation, failed to prevent the inhibition of calmodulin-stimulated (Ca2+ + Mg2+)-ATPase. As dithiothreitol partially restores the enzyme activity, we postulated that hemin interacts with the thiol groups of the enzyme.     

Cholera toxin blocks glucagon-mediated inhibition of the liver plasma membrane (Ca2+-Mg2+)-ATPase

We have previously shown that liver plasma membrane (Ca2+-Mg2+)-ATPase activity is inhibited by glucagon. To investigate the possible involvement of a GTP-binding (G) protein in this regulation, we have examined the effects of pertussis toxin and cholera toxin on inhibition of (Ca2+-Mg2+)-ATPase by glucagon. Treatment of liver plasma membranes with pertussis toxin did not affect the sensitivity of (Ca2+-Mg2+)-ATPase to the hormone. In contrast, treatment of plasma membranes or prior injection of animals with cholera toxin prevented inhibition of the (Ca2+-Mg2+)-ATPase by glucagon. Even though adenylate cyclase activity was increased by cholera toxin treatment, addition of cyclic AMP did not mimic the effect of cholera toxin in blocking glucagon-mediated inhibition of (Ca2+-Mg2+)-ATPase activity. These data suggest that a cholera toxin-sensitive protein, perhaps Gs or a Gs-like protein, is involved in the regulation of liver (Ca2+-Mg2+)-ATPase activity. The results emphasize the possible role of Gs-like proteins in regulation of enzymes other than adenylate cyclase and suggest that the study of (Ca2+-Mg2+)-ATPase may provide a useful enzymatic system to examine such regulation.     

Conformational changes of the nucleotide site of the plasma membrane Ca2+-ATPase probed by fluorescence quenching

Fluorescence quenching by the water-soluble ions I(-) and Cs(+) was used to probe solvent accessibility and polarity of the nucleotide/fluorescein isothiocyanate binding pocket of the purified soluble Ca(2+)-ATPase from plasma membranes. The E(1).Ca.CaM conformer was the least accessible state studied, presenting the lowest suppression constant (K(q)) for both I(-) (K(q) = 6.7 M(-)(1)) and Cs(+) (K(q) = 0.7 M(-)(1)). Accessibility to I(-) was similar for the E(2).VO(4) and E(1).Ca states (K(q) = 7.13 and 7.5 M(-)(1), respectively), whereas E(2) was slightly more accessible (K(q) = 9.1 M(-)(1)). The phosphorylated state E(2)-P presented the highest accessibility, with a K(q) of 16.5 M(-)(1), very near the K(q) of 20.3 M(-)(1) for free FITC. I(-) was unequivocally a better fluorescence quencher, being usually nearly 3-fold as efficient as Cs(+), as indicated by the K(q)(I(-))/K(q)(Cs(+)) ratio (R(q)). The advent of a positive charge cluster on the nucleotide/fluorescein binding pocket in different states was suggested by the increase in R(q), which reached a value as high as 9.5 for the E(1).Ca.CaM conformer. These results indicate (i) a very high water accessibility of the nucleotide/fluorescein pocket for E(2)-P that (ii) is more restricted on the free E(2) state and (iii) becomes rather lower for the E(1).Ca states. Additionally, a positive charge effect of amino acids on the nucleotide site, possibly related to ATP binding and phosphoryl transfer, appears in these E(1).Ca states, being absent in the phosphorylated and nonphosphorylated E(2) states.     

Purification of (Ca2+-Mg2+)-ATPase from rat liver plasma membranes

The Ca2+-stimulated, Mg2+-dependent ATPase from rat liver plasma membranes was solubilized using the detergent polyoxyethylene 9 lauryl ether and purified by column chromatography using Polybuffer Exchanger 94, concanavalin A-Sepharose 4B, and Sephadex G-200. The molecular weight of the enzyme, estimated by gel filtration in the presence of the detergent on a Sephadex G-200 column, was 200,000 +/- 15,000. The enzyme was purified at least 300-fold from rat liver plasma membranes and had a specific activity of 19.7 mumol/mg/min. Polyacrylamide gel electrophoresis under nondenaturing conditions of the purified enzyme indicated that the enzymatic activity correlated with the major protein band. In sodium dodecyl sulfate-polyacrylamide gel electrophoresis, one major band in the molecular weight range of 70,000 +/- 5,000 was seen. The isoelectric point of the purified enzyme was 6.9 +/- 0.2 as determined by analytical isoelectric focusing. The enzyme was activated by Ca2+ with an apparent half-saturation constant of 87 +/- 2 nM for Ca2+. Calmodulin and trifluoperazine at the concentration of 1 microgram/ml and 100 microM, respectively, had no effect on the enzymatic activity.     

Comparison of (Ca2+-Mg2+)-ATPase and Ca2+-ATPase in rat cardiac sarcolemma

The calcium dependency of the Ca2+-pump ATPase of rat cardiac sarcolemma was investigated in the presence and absence of EGTA and EDTA in combination with two free Mg2+-ion concentrations. The results showed: that Mg2+-ions are not essential for the turnover of the Ca2+-pump ATPase; that the Ca2+-affinity is regulated by the concentration of the calcium-chelator complex present in the medium; that (Ca2+-Mg2+)-ATPase and Ca2+-ATPase are probably expressions of the same Ca2+-pump ATPase in the plasma membrane of the cell.     

Conformational differences between the E1 and E2 states of the calcium adenosinetriphosphatase of the erythrocyte plasma membrane as revealed by circular dichroism and fluorescence spectroscopy

Different conformational states of the purified plasma membrane Ca2+-ATPase from pig erythrocytes have been detected by circular dichroism (CD) and fluorescence spectroscopy. The helical content of the enzyme decreased by about 10% in the transition from the Ca2+ high-affinity form (10 microM free Ca2+ = E1 state) to the VO4(3-)-inhibited state (20 microM VO4(3-) = E2 state). The changes in the CD spectra did not show full reversibility upon reversing the E1-E2 transition, whereas those in the fluorescence spectra did. A temperature-dependent loss of alpha-helical content in the presence of Ca2+ was also observed. Intrinsic fluorescence measurements revealed an increase in fluorescence intensity upon addition of Ca2+. The change was fully reversed by ethylene glycol bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid. The increase in fluorescence intensity was partly reversed by adding ATP, an effect which is suggested to correspond to the "Ca2+-occluded" form of the ATPase. The steady-state level of the fluorescence intensity was stable for several minutes in the presence of 100 microM ATP. By contrast, the decrease of fluorescence intensity induced by limiting concentrations of ATP (= 1 microM) was only transient, indicating the decomposition of the phosphorylated intermediate of the ATPase and the reestablishment of the Ca2+ high-affinity form of the enzyme.     

Long-term stabilization and crystallization of (Ca2+ + Mg2+)-ATPase of detergent-solubilized erythrocyte plasma membrane.

Conditions which were optimal for the stabilization of Ca2(+)-transporting ATPase in solubilized sarcoplasmic reticulum membranes (Piku?la, S., Mullner, N., Dux, L. and Martonosi, A. (1988) J. Biol. Chem. 263, 5277-5286) were also found conducive for preservation of (Ca2+ + Mg2+)-ATPase activity in detergent-solubilized erythrocyte plasma membrane for up to 60 days. Of particular importance for the stabilization of calmodulin-stimulated Ca2(+)-dependent activity of (Ca2+ + Mg2+)-ATPase of solubilized erythrocyte plasma membrane was the presence of Ca2+ (10-20 mM), glycerol, anti-oxidants, proteinase inhibitors and appropriate detergents. Among eight detergents tested octaethylene glycol dodecyl ether, polyoxyethylene glycol(10) lauryl alcohol and polydocanol were found to be promotive in long-term preservation of the enzyme activity. Under these conditions (Ca2+ + Mg2+)-ATPase of erythrocyte ghosts became highly stable and developed microcrystalline arrays after storage for 35 days. Electron micrographs of the negatively stained and thin sectioned material indicated that crystals of purified, detergent-solubilized, lipid-stabilized erythrocyte (Ca2+ + Mg2+)-ATPase differ from those of Ca2(+)-ATPase of detergent-solubilized sarcoplasmic reticulum microsomes.     

Decrease of apparent calmodulin affinity of erythrocyte (Ca2+ + Mg2+)-ATPase at low Ca2+ concentrations

The calmodulin activation of the (Ca2+ + Mg2+)-ATPase (ATP phosphohydrolase, EC 3.6.1.3) in human erythrocyte membranes was studied in the range of 1 nM to 40 microM of purified calmodulin. The apparent calmodulin-affinity of the ATPase was strongly dependent on Ca2+ and decreased approx. 1000-times when the Ca2+ concentration was reduced from 112 to 0.5 microM. The data of calmodulin (Z) activation were analyzed by the aid of a kinetic enzyme model which suggests that 1 molecule of calmodulin binds per ATPase unit and that the affinities of the calcium-calmodulin complexes (CaiZ) decreases in the order of Ca3Z greater than Ca4Z greater than Ca2Z greater than or equal to CaZ. Furthermore, calmodulin dissociates from the calmodulin-saturated Ca2+-ATPase in the range of 10(-7)-10(-6) M Ca2+, even at a calmodulin concentration of 5 microM. The apparent concentration of calmodulin in the erythrocyte cytosol was determined to be 3 to 5 microM, corresponding to 50-80-times the cellular concentration of Ca2+-ATPase, estimated to be approx. 10 nmol/h membrane protein. We therefore conclude that most of the calmodulin is dissociated from the Ca2+-transport ATPase in erythrocytes at the prevailing Ca2+ concentration (probably 10(-7)-10(-8) M) in vivo, and that the calmodulin-binding and subsequent activation of the Ca2+-ATPase requires that the Ca2+ concentration rises to 10(-6)-10(-5) M.     

Allosteric modulation of Leishmania donovani plasma membrane Ca(2+)-ATPase by endogenous calmodulin.

The plasma membrane of the human pathogen Leishmania donovani possesses a high-affinity transmembrane Ca(2+)-ATPase that has its catalytic site oriented toward the cytoplasmic milieu (Ghosh, J., Ray, M., Sarkar, S., and Bhaduri, A. (1990) J. Biol. Chem. 265, 11345-11351). When the enzyme is studied in its more authentic, physiologically relevant, membrane-associated form, it exhibits pronounced sigmoidal kinetics with Ca2+ (K0.5 approximately 700 nM) in a trans-1,2-diaminocyclohexane-N,N,N',N'-tetraacetic acid buffering system that effectively complexes all available Mg2+. Addition of exogenous Mg2+ (60 microM) completely abolishes sigmoidicity and establishes strictly hyperbolic kinetics, and the Km for Ca2+ reduces to 100 nM. Mg2+ can be replaced by heterologous calmodulin. The exclusive dependence of the enzyme on only Ca2+ for its activity and its positive allosteric modulation by Mg2+ distinguish this enzyme from other well-characterized plasma membrane Ca(2+)-ATPases. Employing this Ca(2+)-ATPase as the assay system, a soluble endogenous activating protein factor was purified that, by several criteria, corresponds to authentic calmodulin. The parasite calmodulin shifts the kinetics to hyperbolic kinetics, increases the Vmax 2-fold, and most important lowers the Km (approximately 100 nM) to a physiological level. The interaction with endogenous calmodulin thus converts the enzyme from a totally inactive to a fully active state.     

The inhibitor of liver plasma membrane (Ca2+-Mg2+)-ATPase. Purification and identification as a mediator of glucagon action

Rat liver plasma membranes contain (Ca2+-Mg2+)-ATPase sensitive to inhibition by both glucagon and Mg2+. We have previously shown that Mg2+ inhibition is mediated by a 30,000-dalton inhibitor, originally identified as a membrane-bound protein. In fact, this inhibitor is also present in the 100,000 X g supernatant of the total liver homogenate. Its purification was achieved from this fraction by a combination of ammonium sulfate washing, gel filtration, and cationic exchange chromatography. N-Ethylmaleimide (NEM) treatment caused the inactivation of the purified inhibitor, which suggested that this protein possesses at least one NEM-sensitive sulfhydryl group essential for its activity. Treatment of the liver plasma membranes with NEM resulted in a 2- and 5-fold decrease in the affinity of the (Ca2+-Mg2+)-ATPase for glucagon and Mg2+, respectively, while the basal enzyme activity remained unchanged. This effect of NEM was concentration-, pH-, and time-dependent, optimal conditions being obtained by a 60-min treatment of plasma membranes with 50 mM NEM, at pH 7 and at 4 degrees C. The presence of 0.5 mM Mg2+ during NEM treatment of the plasma membranes prevented NEM inactivation. Reconstitution experiments showed that addition of the purified inhibitor to NEM-treated plasma membranes restored the inhibitions of the (Ca2+-Mg2+)-ATPase by both magnesium and glucagon. It is proposed that the (Ca2+-Mg2+)-ATPase inhibitor not only confers its sensitivity of the liver (Ca2+-Mg2+)-ATPase to Mg2+, but also mediates the inhibition of this system by glucagon.     

Conformational changes in erythrocyte membranes by prostaglandins as measured by circular dichroism

Membranes were prepared from fresh, washed human erythrocytes by hemolysis and washing with 5 mm sodium phosphate buffer (pH 7.4). The mean residue ellipticity, [θ], of erythrocyte membrane circular dichroism was altered by prostaglandin E₁ or prostaglandin F_2α at 37 °C when observed from 250 nm to 190 nm. The decrease in negativity of [θ] with 10^?6m prostaglandin E₁ was 12.7% at 222 nm and 17.7% at 208 nm, and with 10^?6m prostaglandin F_2α 22.5% and 34.2%, respectively (P < 0.01). Similar changes in [θ] were observed at lower concentrations of prostaglandins. No strict relationship between amount of change of [θ] and prostaglandin concentrations of 3 × 10^?5m to 3 × 10^?12m was evident. A persistent alteration of [θ] with prostaglandin was observed at 37 °C. Transient change of [θ] occurred at 25 °C with prostaglandin. No change of [θ] was observed at 15 or 20 °C. Buffer or palmitic acid were without effect on membrane [θ]. Phosphatidyl inositol or methyl arachidonate caused an increase in negativity of membrane spectra. The observed alterations of membrane [θ] did not arise from changes in light scattering as the OD₇₀₀–OD₂₀₀ of membranes was not changed by prostaglandin. Effects of prostaglandin were not dependent on light path length. The prostaglandin E₁ antagonist, 7-oxa-13-prostynoic acid, at 10^?7m produced no change of [θ] of membrane spectra and prevented the otherwise demonstrable effects of 10^?10m prostaglandin E₁ on [θ]. The decrease in negativity of [θ] at 222 nm is indicative of a decrease in ellipticity of membrane protein. These studies suggest that prostaglandins may act by inducing a conformational change in membrane protein.     

Human erythrocyte Ca2+-Mg2+-ATPase: mechanism of stimulation by Ca2+.

We have critically evaluated hydrodynamic data from 21 proteins whose molecular dimensions are known from X-ray crystallography. We present two useful equations relating the molecular weights and sedimentation coefficients of globular proteins. The hydrodynamic data combined with data for small molecules from the literature indicate that failure of the Stokes equation occurs only for molecular weights <850. Calculated hydration values for the 21 proteins have a mean value and standard deviation of 0.53 ± 0.26 g H₂O/g protein. Furthermore, statistical arguments indicate that only 5.3% of the variance is due to experimental error. The mean value and especially the dispersion of values are in sharp contrast to the values 0.36 ± 0.04 obtained by others from nmr measurements on frozen protein solutions. Hydration values calculated from nmr measurements are closely correlated with the number of charged and polar amino acid residues. In contrast to this result, our analysis of the amino acid compositions of the four proteins with the lowest hydration and the four monomeric proteins with the highest shows that the range of values we observe cannot be accounted for on the basis of amino acid composition. In fact there appears to be a weak correlation between the number of apolar residues and hydrodynamic hydration. We therefore conclude that the dispersion must result from variations in fine details of the surface structures of individual proteins. We propose a model of hemispherical clathrate cages which if correct, would account for the differences in the data obtained by these two methods.     

A protein modulator of erythrocyte membrane (Ca2+ + Mg2+)-ATPase inhibitor protein

A protein modulator of erythrocyte membrane (Ca²⁺ + Mg²⁺)-ATPase inhibitor protein was purified to apparent homogeneity from pig membrane-free hemolysate by a combination of carboxymethyl-Sephadex chromatography, gel filtration, chromatofocusing (pH 7-4) and subsequent removal of trace inhibitor protein by salt treatment. Gel filtration gave a molecular weight of 57 500 for the purified protein modulator, while SDS-polyacrylamide gel electrophoresis of dithiothreitol-treated modulator revealed one single band with a molecular weight of 29 000. Isoelectric focusing of the dithiothreitol-treated protein revealed one band (isoelectric pH 4.85), while untreated modulator gave an extra band (isoelectric pH 4.96). It contains no methionine and has an acidic content 73% higher than that of its basic residues. Freshly prepared or dithiothreitol-treated modulator suppressed both pig and human erythrocyte (Ca²⁺ + Mg²⁺)-ATPase inhibitor protein activity, but did not affect ATPase and calmodulin activities. Modulator-coupled Affi-Gel 15 could be employed for purification of the protein inhibitor.