A B S T R A C T Experiments were designed to characterize several partial reactions of the Na,K-ATPase and to demonstrate that a model can be defined that reproduces most of the transport features of the pump with a single set of kinetic parameters. We used the fluorescence label 5-iodoacetamidofluorescein, which is thought to be sensitive to conformational changes, and the styryl dye RH 421, which can be applied to detect ion-binding and -release reactions. In addition transient electric currents were measured, which are associated mainly with the E l ~ E2 conformational transition. Numerical simulations were performed on the basis of a reaction model, that has been developed from the Post-Albers cycle. Analysis of the experimental data allows the determination of several rate constants of the pump cycle. Our conclusions may be summarized as follows: (a) binding of one Na + ion at the cytoplasmic face is electrogenic. This Na + ion is specifically bound to a neutral binding site with an affinity of 8 mM in the presence of 10 mM Mg 2+. In the absence of divalent cations, the intrinsic binding affinity was found to be 0.7 mM. (b) The analysis of fluorescence experiments with the cardiotonic steroid strophanthidin indicates that the 5-iodoacetamidofluorescein label monitors the conformational transition (Na3)EI-P ---, P-Ez(Na2), which is accompanied by the release of one Na + ion. 5-IAF does not respond to the release of the subsequent two Na + ions, which can be monitored by the RH 421 dye. These experiments indicate further that the conformational transition E,P ~ P-E2 is the rate limiting process of the Na + translocation. The corresponding rate constant was determined to be 22 s -1 at 20~ From competition experiments with cardiotonic steroids, we estimated that the remaining 2 Na § ions are released subsequently with a rate constant of at least 5,000 s -1 from their negatively charged binding sites. (c) Comparing the fluorescence experiments with electric current transients, which were performed at various Na concentrations in the absence and presence of strophanthidin, we found that the transition (Na3)'EI-P --, P-E2"(Na2) is the major charge translocating step in the reaction sequence Na3'El ~ (Na3)'E1-P ~ P-E2"(Na~) ~ P-E 2. The subsequent release of 2 Na + ions contributed less than 25% to the total electric current transient. 198 THE JOURNAL OF GENERAL PHYSIOLOGY 9 VOLUME 104 9 1994 binding can be explained by a kinetic model. A quantitative description has been obtained under the assumption that these inhibitors bind only to the states P-E~(Na2) and P-E~ (K~). (e) Most of our experiments can be described by a modified Post-Albers scheme. A set of the kinetic parameters in this scheme has been determined by the experiments presented or by data from literature. Numerical simulations using this set are consistent with the presented data.

A charge-pulse technique was designed to measure charge movements in the Na-transport mode of the Na,K-ATPase in membrane fragments adsorbed to a planar lipid bilayer with high time resolution. 1) Na+ transport was measured as a function of membrane potential, and 2) voltage-dependent extracellular ion binding and release were analyzed as a function of Na+ concentration and membrane potential. The results could be fitted and explained on the basis of a Post-Albers cycle by simulations with a mathematical model. The minimal reaction sequence explaining the electrogenicity of the pump consists of the following steps: (Na3)E1-P <--> P-E2(Na3) <--> P-E2(Na2) <--> P-E2(Na) <--> P-E2. The conformational change, E1 to E2, is electrogenic (beta 0 < or = 0.1) and the rate-limiting step of forward Na+ transport with a rate constant of 25 s-1 (T = 20 degrees C). The first ion release step, P-E2(Na3) <--> P-E2(Na2), is the major charge translocating process (delta 0 = 0.65). It is probably accompanied by a protein relaxation in which the access structure between aqueous phase and binding site reduces the dielectric distance. The release of the subsequent Na+ ions has a significantly lower dielectric coefficient (delta1 = delta 2 = 0.2). Compared with other partial reactions, the ion release rates are fast (1400 s-1, 700 s-1, and 4000 s-1). On the basis of these findings, a refined electrostatic model of the transport cycle is proposed.

Nonstationary electric currents are described which are generated by the Na,K-pump. Flat membrane sheets 0.2-1 micron in diameter containing a high density of oriented Na,K-ATPase molecules are bound to a planar lipid bilayer acting as a capacitive electrode. In the aqueous phase adjacent to the bound membrane sheets, ATP is released within milliseconds from an inactive, photolabile precursor ("caged" ATP) by an intense flash of light. After the ATP-concentration jump, transient current and voltage signals can be recorded in the external circuit corresponding to a translocation of positive charge across the pump protein from the cytoplasmic to the extracellular side. These electrical signals which can be suppressed by inhibitors of the Na,K-ATPase require the presence of Na+ but not of K+ in the aqueous medium. The intrinsic pump current Ip(t) can be evaluated from the recorded current signal, using estimated values of the circuit parameters of the compound membrane system. Ip(t) exhibits a biphasic behavior with a fast rising period, followed by a slower decline towards a small quasi-stationary current. The time constant of the rising phase of Ip(t) is found to depend on the rate of photochemical ATP release. Further information on the microscopic origin of the current transient can be obtained by double-flash experiments and by chymotrypsin modification of the protein. These and other experiments indicate that the observed charge-translocation is associated with early events in the normal transport cycle. After activation by ATP, the pump goes through the first steps of the cycle and then enters a long-lived state from which return to the initial state is slow.

No abstract

Abstract. When purple-membrane fragments from Halobacterium halobium are added to one aqueous phase of a positively-charged black lipid membrane, the membrane becomes photoelectrically active. Under normal conditions the steady-state photo-current is extremely low, but increases considerably when the lipid bilayer is doped with proton-permeable gramicidin channels or with a lipophilic acid-base system. These findings indicate that the purple-membrane sheets are bound to the surface of the bilayer, forming a sandwich-like structure. The time-behaviour of the photocurrent may be interpreted on the basis of a simple equivalent circuit which contains the conductance and capacitance of the purple membrane in series with the conductance and capacitance of the lipid bilayer. From the dependence of the photocurrent on the polarization of the exciting light the average angle between the transition moment of the retinal chromophore and the plane of the bilayer was calculated to be about 28 degrees. Furthermore, it was shown that chromophore-free apomembrane binds to the lipid bilayer and that its photoelectrical activity can be restored in situ by adding all-trans-retinal to the aqueous phase.

Among the major obstacles to pharmacological and structural studies of integral membrane proteins (MPs) are their natural scarcity and the difficulty in overproducing them in their native form. MPs can be overexpressed in the non-native state as inclusion bodies, but inducing them to achieve their functional three-dimensional structure has proven to be a major challenge. We describe here the use of an amphipathic polymer, amphipol A8-35, as a novel environment that allows both beta-barrel and alpha-helical MPs to fold to their native state, in the absence of detergents or lipids. Amphipols, which are extremely mild surfactants, appear to favor the formation of native intramolecular protein-protein interactions over intermolecular or protein-surfactant ones. The feasibility of the approach is demonstrated using as models OmpA and FomA, two outer membrane proteins from the eubacteria Escherichia coli and Fusobacterium nucleatum, respectively, and bacteriorhodopsin, a light-driven proton pump from the plasma membrane of the archaebacterium Halobacterium salinarium.

In the first part of the paper, evidence has been presented that electrochromic styryl dyes, such as RH 421, incorporate into Na,K-ATPase membranes isolated from mammalian kidney and respond to changes of local electric field strength. In this second part of the paper, fluorescence studies with RH-421-labeled membranes are described, which were carried out to obtain information on the nature of charge-translocating reaction steps in the pumping cycle. Experiments with normal and chymotrypsin-modified membranes show that phosphorylation by ATP and occlusion of Na+ are electroneutral steps, and that release of Na+ from the occluded state to the extracellular side is associated with translocation of charge. Fluorescence signals observed in the presence of K+ indicate that binding and occlusion of K+ at the extracellular face of the pump is another major electrogenic reaction step. The finding that the fluorescence signals are insensitive to changes of ionic strength leads to the conclusion that the binding pocket accommodating Na+ or K+ is buried in the membrane dielectric. This corresponds to the notion that the binding sites are connected with the extracellular medium by a narrow access channel ("ion well"). This notion is further supported by experiments with lipophilic ions, such as tetraphenylphosphonium (TPP+) or tetraphenylborate (TPB-), which are known to bind to lipid bilayers and to change the electrostatic potential inside the membrane. Addition of TPP+ leads to a decrease of binding affinity for Na+ and K+, which is thought to result from the TPP(+)-induced change of electric field strength in the access channel.

Experiments with large unilamellar dioleoylphosphatidylcholine vesicles were carried out in order to study the effect of membrane potential on the fluorescence of Oxonol VI. A partition equilibrium of dye between membrane and water was found to exist with a partition coefficient Y=CJjpid/cwater of about 19000 (at zero voltage). In the presence of an inside-positive membrane potential, the negatively charged dye accumulates in the intravesicular aqueous space according to a Nernst equilibrium. This leads to an increased adsorption of dye to the inner lipid monolayer and to a concomitant increase of fluorescence. The fluorescence change can be calibrated as a function of transmembrane voltage by generating a potassium diffusion potential in the presence of valinomycin. The intrinsic fluorescence of the membrane-bound dye is not affected by voltage; the whole influence of voltage on the fluorescence results from voltage-dependent partitioning of the dye between water and membrane. The voltage dependence of the apparent partition coefficient can be quantitatively described by a three-capacitor model in which the dye is assumed to bind to adsorption planes located on the hydrocarbon side of the membrane / solution interface. Oxonol VI was found to be suitable for detecting changes of membrane potential associated with the activity of the (Na+ + K+)-ATPase in reconstituted vesicles. When ATP is added to the external medium, pump molecules with the ATP-binding side facing outward become activated; this results in a translocation of net positive charge towards the vesicle interior. Under this condition, fluorescence changes corresponding to (inside-positive) potentials of up to 150-200 mV are observed. After the build-up of the membrane potential, a quasi-stationary state is reached in which the pump current is compensated by a back-flow of charge through passive conductance pathways.