Functional characterization of reconstituted sarcoplasmic reticulum vesicles

A detailed functional characterization of reconstituted sarcoplasmic reticulum (SR) vesicles with similar lipid content as normal SR was obtained by studies of ATPase activity and calcium transport in transient state, steady state, and equilibrium conditions. For this purpose, enzyme phosphorylation with ATP, hydrolytic activity, calcium transport, phosphorylation with Pi, and ATP synthesis by reversal of the pump were measured, and utilized to demonstrate function and orientation of catalytic sites. The preparations used in these studies displayed the highest activity reported for reconstituted sarcoplasmic reticulum systems. The rates of phosphoenzyme formation from ATP and hydrolysis as well as steady state levels matched the values obtained with normal SR vesicles. Calcium transport and repeated cycles of ATP synthesis by reversal of the pump were also obtained. However, the efficiency of transport and ATP synthesis from a Ca2+ gradient was approximately three times lower than in native vesicles. This deficiency could not be attributed to passive calcium leak from the reconstituted vesicles but, in part, can be explained by the bidirectional alignment of the calcium pump in reconstituted SR. It is suggested that vectorial transport requires a more complex level of protein structure than that for sustaining simple ATPase activity. Time resolution of the phosphorylation reaction by rapid quench methods can be used to estimate the orientation of the calcium pump in the membrane. Such studies indicate that the calcium pump protein is largely bidirectionally oriented in reconstituted SR vesicles.     

Comparison of the profile structures of isolated and reconstituted sarcoplasmic reticulum membranes

The profile structures of functional reconstituted sarcoplasmic reticulum (RSR) membranes were investigated as a function of the lipid/protein (L/P) ratio via x-ray diffraction studies of hydrated oriented multilayers of these membranes to a resolution of 10-15 A, and neutron diffraction studies on these multilayers to lower resolutions. Our results at this stage of investigation indicate that reconstitution of SR with variable amounts of Ca2+ pump protein for L/P ratios greater than 88 results in closed membraneous vesicles in which the Ca2+ pump protein is distributed asymmetrically in the membrane profile; a majority of the protein density is contained primarily in the extravesicular half of the membrane profile whereas a relatively lesser portion of the protein spans the hydrocarbon core of the RSR membranes. These RSR membranes are functionally similar and resemble isolated light sarcoplasmic reticulum in both profile structure and function at a comparable L/P ratio. Reconstitution with greater amounts of Ca2+ pump protein (e. g. L/P approximately 50-60) resulted in substantially less functional membranes with a dramatically thicker profile structure.     

Characterization of a chloride channel reconstituted from cardiac sarcoplasmic reticulum

We have characterized a voltage-sensitive chloride channel from cardiac sarcoplasmic reticulum (SR) following reconstitution of porcine heart SR into planar lipid bilayers. In 250 mm KCl, the channel had a main conductance level of 130 pS and exhibited two substrates of 61 and 154 pS. The channel was very selective for Cl^– over K⁺ or Na⁺ ( and ). It was permeable to several anions and displayed the following sequence of anion permeability: SCN^– > I^– > NO ₃ ^– Br^– > Cl^– > f^– > HCOO^–. Single-channel conductance saturated with increasing Cl^– concentrations (K _m= 900 mm and _max = 488 pS). Channel activity was voltage dependent, with an open probability ranging from 1.0 around 0 mV to 0.5 at +80 mV. From –20 to +80 mV, channel gating was time-independent. However, at voltages below –40 mV the channel entered a long-lasting closed state. Mean open times varied with voltage, from 340 msec at –20 mV to 6 msec at +80 mV, whereas closed times were unaffected. The channel was not Ca²⁺-dependent. Channel activity was blocked by disulfonic stilbenes, arylaminobenzoates, zinc, and cadmium. Single-channel conductance was sensitive to trans pH, ranging from 190 pS at pH 5.5 to 60 pS at pH 9.0. These characteristics are different from those previously described for Cl^– channels from skeletal or cardiac muscle SR.We thank Dr. Barry Pallotta for help with open and closed intervals analysis and Dr. Gerhard Meissner for his suggestions for the preparation of cardiac sarcoplasmic reticulum membranes. This work was supported by a grant from the National Institutes of Health to R.L.R. and a Student Grant-in-Aid from the American Heart Association, North Carolina affiliate to C.T. R.L.R. is an Established Investigator of the American Heart Association.     

Temperature-induced transitions of function and structure in sarcoplasmic reticulum membranes

A transition in the temperature dependences of Ca²⁺ accumulation and ATPase activity occurs at 20 ° C in Sarcoplasmic reticulum membranes. The transition is characterized by an abrupt change in the activation energies for the cation transport process and the associated enzyme activities. The difference in activation energies below and above 20 °C appears to be due to changes in the entropy of activation rather than in the free energy of activation. Also, the temperature dependences of spectral parameters of lipophilic spin-labeled probes and protein-bound spin labels exhibit different behaviors on either side of this temperature. Above 20 °C the lipid matrix probed by the labels exhibits a large increase in molecular motion and a decrease in the apparent ordering of lipid alkyl chains. In addition, labels covalently bound to enzymic reactive sites indicate that the motion of protein side-chains is sensitive to this transition. The results are consistent with an order-disorder transition involving the lipid alkyl chains of the Sarcoplasmic membrane, and with a model in which molecular motion, Ca²⁺ transport and enzyme activity are limited by local viscosity of hydrophobic regions at temperatures below the transition.Another modification of the Sarcoplasmic reticulum membrane occurs between 37 and 40 °C. It appears that at this temperature the processes governing Ca²⁺ accumulation and ATPase activity are uncoupled, and Ca²⁺ accumulation is inhibited, while ATPase activity and passive Ca²⁺ efflux proceed at rapid rates. Parallel transitions of spectroscopic parameters originating from spin labels, covalently bound to the Sarcoplasmic reticulum ATPase, indicate that the uncoupling is due to a thermally-induced protein conformational change.     

Study of the electrokinetic properties of reconstituted sarcoplasmic reticulum vesicles

A study of electrokinetic properties of reconstituted sarcoplasmic reticulum was undertaken to determine the nature of the groups bearing the negative charge of the membrane. After incorporation of phosphatidylcholine into the bilayer, it was found that the Ca2+-ATPase embedded in functional vesicles bore 3e- per mole. When the surface charge density of the hydrodynamic particles became more negatively charged by incorporation of phosphatidylserine molecules, the reconstituted vesicles had a tendency to build large structures resulting from vesicle-vesicle interaction and containing large amounts of divalent cations. These aggregated structures may partially explain the discrepancy observed between the expected value of the surface charge density and the data obtained by electrophoretic mobility measurements. This work emphasizes the importance of a renewal of the classical interpretation of electrophoretic mobility data in order to analyze the results obtained with biological material. To explain the energy transduction process which takes place in the sarcoplasmic reticulum membrane, it was of interest to determine whether or not variations of the surface electrical properties affect the calcium ion translocation upon ATP hydrolysis. Relatively significant modifications of the bilayer composition and surface charge density did not appreciably affect the calcium transport activity.     

Chemical modification of sarcoplasmic reticulum membranes

The usefulness of chemical cross-linking and ¹²⁵I-labeling techniques in the analysis of protein-protein interactions and membrane polarity was evaluated on sarcoplasmic reticulum membranes. Treatment of fragmented sarcoplasmic reticulum vesicles with glutaraldehyde, dimethylsuberimidate, or copper-phenanthroline leads to the formation of high molecular weight aggregates of the Ca²⁺ transport ATPase; intermediate polymers of functionally and structurally interesting sizes accumulated only occasionally and in amounts of questionable significance. Coupling of membrane proteins with tolylene 2,4-diisocyanate-albumin inhibited tht ATPase activity and caused the appearance of high molecular weight aggregates and a band of about 160 000 dalton which corresponds to the ATPase-albumin complex.Even after the 100 000 dalton band of the Ca²⁺-transport ATPase was severely diminished by cross-linking with copper-phenanthroline or toluene diisocyanate-albumin, the Ca²⁺ binding proteins of sarcoplasmic reticulum remained unreacted. A consistent finding was the presence of dimers of the Ca²⁺ transport ATPase in aged preparations of sarcoplasmic reticulum which were converted upon reduction with β-mercaptoethanol into 100 000 dalton units.Microsomes were labeled with ¹²⁵I in the presence of lactoperoxidase, glucose oxidase, and glucose and the radioactivity oft he various protein components was measured after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The specific activity of calsequestrin was many times greater than that of the Ca⁺ transport ATPase suggesting that it is exposed on the outside surface may be sterically hindered from access by bulky reagents (tolylene diisocyanate-albumin, ferritin-labeled anti-calsequestrin antibodies, proteolytic enzymes, etc.), as calsequestin becomes highly reactive with these agents only after its release from the membrane.     

Ca 2+ uptake in reconstituted sarcoplasmic reticulum vesicles

The reconstitution of functional sarcoplasmic reticulum vesicles capable of Ca²⁺ transport has been achieved. Sarcoplasmic reticulum vesicles are first solubilized with deoxycholate and then reassembled into membranous vesicles by removal of the detergent using dialysis. The Ca²⁺ pump protein can, by itself, be reconstituted to form membranous vesicles capable of energized Ca²⁺ binding and uptake. The lipid content of the reconstituted vesicles is about the same as that of the original sarcoplasmic reticulum vesicles. The reconstituted vesicles have an elevated ATPase activity. Ca²⁺ binding and uptake in the presence of ATP are restored to about 25% and 50%, respectively.     

ATP-dependent phosphate transport in sarcoplasmic reticulum and reconstituted proteoliposomes

During uptake of Ca2+ by rabbit sarcoplasmic reticulum, about 1 mumol of 32Pi was taken up per mumol 45Ca2+ transported. The uptake of Pi was dependent on external Ca2+, Mg2+ and ATP. Intravesicular Ca2+ did not substitute for external Ca2+. In contrast to the accumulation of Ca2+ which was abolished by the ionophore A23187, the uptake of Pi continued to take place provided sufficient Ca2+ was present in the medium. Thus, a Ca2+ gradient did not seem to be required. Similar observations were made with proteoliposomes reconstituted with membrane preparations of sarcoplasmic reticulum and soybean phospholipids. However, when purified Ca2+ -ATPase was used for reconstitution, there was ATP-dependent Ca2+ uptake but no ATP-dependent Pi transport was observed. These data show that the mechanism of Pi transport cannot be a passive movement in response to a Ca2+ gradient but appears to be catalyzed by a specific protein, which is inactivated during purification of the Ca2+ -ATPase. A protein that catalyzes Pi transport in reconstituted vesicles has been solubilized by extraction of sarcoplasmic reticulum with sodium cholate.     

Raman spectroscopic investigations of sarcoplasmic reticulum membranes

Raman spectra are presented for sarcoplasmic reticulum membranes. Interpretation of the 1000–1130 cm^?1 region of the spectrum indicates that the sarcoplasmic reticulum membrane may be more fluid than erythrocyte membranes that have been examined by the same technique. The fluidity of the membrane also manifests itself in the amide I portion of the membrane spectrum with a strong 1658 cm^?1 band characteristic of CC stretching in hydrocarbon side chains exhibiting cis conformation. This band is unaltered in intensity and position in H₂O and in ²H₂O thus obscuring amide I protein conformation. Of particular interest is the appearance of strong, resonantly enhanced bands at 1160 and 1527 cm^?1 attributable to membrane-associated carotenoids.     

Electron density levels of sarcoplasmic reticulum membranes

Low-angle X-ray diffraction has been recorded from oriented preparations of sacroplasmic reticulum membranes in fluid media containing glycerol solutions in different concentrations. Discrete diffraction orders of a lamellar repeat distance ranging from 200 Å to 250 Å have been recorded. Fourier synthesis at a resolution of 17 Å for 0, 10, 20, and 30% glycerol-treated sarcoplasmic reticulum membranes are described. An electron density scale in electrons/$\text{A}\text{?}{}^3$ for these Fourier syntheses has been determined. The question of the correctness of our asymmetric electron density profile for the sarcoplasmic reticulum membrane is critically examined. A study is made on the choice of phases and on the method used to process the X-ray intensities.     

Multilayer planar membranes of sarcoplasmic reticulum.

Multilayer planar membranes were constructed between a pair of cellulose sheets from fragmented sarcoplasmic reticulum (FSR) as well as a mixture of egg yolk lecithin and the Ca2+-ATPase purified from FSR. Since sodium deoxycholate was used instead of organic solvents in order to dissolve phospholipids in the process of the membrane preparation, the total activity of the Ca2+-ATPase was still preserved in the planar membrane of FSR. It was also indicated using a spin label technique that the orientation of phospholipids in the planar membrane of FSR was considerably disturbed by the presence of proteins such as the Ca2+-ATPase included in FSR.     

Correlation of ultrastructure of reconstituted sarcoplasmic reticulum membrane vesicles with variation in phospholipid to protein ratio.

We have previously described the reconstitution of functional membrane vesicles with lipid content similar to that of the normal sarcoplasmic reticulum membrane (approximately 1.0 mumol of phospholipid/mg of protein). The present study describes methodology to prepare reconstituted membrane vesicles with defined phospholipid to protein ratio, both lower and higher than that of the original membrane. The Ca2+ loading rate and efficiency are greatest in the membranes of highest protein content (0.38 mumol of phospholipid/mg of protein), decline slowly as the lipid content is quadrupled, and decrease markedly as the lipid content is quadrupled again. Such membranes of defined composition can be used to study lipid-protein interaction and to correlate membrane structure with composition. The number of particles observed by freeze-fracture electron microscopy can be correlated with protein content, whereas the percentage of smooth domain is proportional to the lipid content of the reconstituted membrane. Since 90% or more of the protein of the reconstituted membrane is the calcium pump protein, the number of particles observed by freeze-fracture is directly proportional to the amount of calcium pump protein in the membrane. The number of pump molecules calculated to be in the membrane is greater by a factor of two than the number of particles which we observed. This multiplicity ratio could be greater depending upon the assumptions made regarding the width of the membrane (see "Appendix"). Thus, it would appear that the particles consist of two or more molecules of pump protein. The change in protein concentration of the membrane is reflected also in thin sections and by negative staining. In thin sections, the broad inner and outer 70 A bands become discontinuous and patchy and, in the limit, approach a symmetrical 20,20,20 A trilayer as the protein content of the membrane becomes small. In an analogous fashion, the concentration of particles at the surface of the membrane, observed by negative staining, decreases with increasing lipid concentration in the membrane. Thus, the correlation of composition with structure can be observed by each of the three methods of sample preparation for electron microscopic analysis.     

Protein conformational transitions in sarcoplasmic reticulum membranes

Fourier-transform infrared spectroscopic studies of sarcoplasmic reticulum proteins, in H2O and D2O, suggest that 10 mM ATP induces a conformational change in those proteins, increasing their contents in alpha-helical and beta-antiparallel structures. Ca2+ on the contrary, is seen to reduce the proportion of alpha-helix and increase the contribution of random coil.     

Sarcoplasmic reticulum. IX. The permeability of sarcoplasmic reticulum membranes

Fragmented sarcoplasmic reticulum (FSR) membranes isolated from rabbit skeletal muscle are impermeable to inulin-¹⁴C (mol wt 5,000), and dextran-¹⁴C (mol wt 15,000–90,000) at pH 7.0–9.0, yielding an excluded space of 4–5 µl/mg microsomal protein. In the same pH range urea and sucrose readily penetrate the FSR membrane. EDTA or EGTA (1 mM) increased the permeability of microsomes to inulin-¹⁴C or dextran-¹⁴C at pH 8–9, parallel with the lowering of the FSR-bound Ca⁺⁺ content from initial levels of 20 nmoles/mg protein to 1–3 nmoles/mg protein. EGTA was as effective as EDTA, although causing little change in the Mg⁺⁺ content of FSR. The permeability increase caused by chelating agents results from the combined effects of high pH and cation depletion. As inulin began to penetrate the membrane there was an abrupt fall in the rate of Ca⁺⁺ uptake and a simultaneous rise in ATPase activity. At 40°C inulin penetration occurred at pH 7.0 with 1 mM EDTA and at pH 9.0 without EDTA, suggesting increased permeability of FSR membranes. This accords with the higher rate of Ca⁺⁺ release from FSR at temperatures over 30°C. The penetration of microsomal membranes by anions is markedly influenced by charge effects. At low ionic strength and alkaline pH acetate and Cl are partially excluded from microsomes when applied in concentrations not exceeding 1 mM, presumably due to the Donnan effect. Penetration of microsomal water space by acetate and Cl occurs at ionic strengths sufficiently high to minimize charge repulsions.