Excess electrical noise during current flow through porous membranes separating ionic solutions.

Spectral analysis of electrical noise from various artificial membrane systems suggests that excess noise of an f-n spectral form, where n is approximately unity, is not primarily a bulk phenomenon simply dependent on the number of charge carriers. Measurements from aqueous and nonaqueous electrolytic resistors, comprised of several different ionic species, show only flat power density spectra under applied currents, even at extreme dilutions. Excess noise of f-n form is observed under applied d-c current in single pore membranes, as previously reported, but is also seen in multipore and polymer mesh membranes. Calculations based on single pore membrane noise data are in significant variance with the bulk charge carrier model proposed by Hooge. These observations suggest that such excess noise occurs in conjunction with anisotropic constraints to ion flow.     

The mitochondrial megachannel is the permeability transition pore

Single-channel electrophysiological recordings from rat liver mitoplast membranes showed that the 1.3-nS mitochondrial megachannel was activated by Ca⁺⁺ and inhibited by Mg⁺⁺, Cyclosporin A, and ADP, probably acting at matrix-side sites. These agents are known to modulate the so-called mitochondrial permeability transition pore (Gunter, T. E., and Pfeiffer, D. R. (1990)Am. J. Physiol. 258, C755–C786) in the same manner. Furthermore, the megachannel is unselective, and the minimum pore size calculated from its conductance is in agreement with independent estimates of the minimum size of the permeabilization pore. The results support the tentative identification of the megachannel with the pore believed to be involved in the permeabilization process.Abbreviations used: PT: permeability transition; PTP: permeability transition pore; MMC: mitochondrial megachannel; IMAC: inner membrane anion channel. P_A: permeability of ion A. CSP: Cyclosporin A.     

Diffusion and 1/f noise

Summary The noise associated with ion transport through porous membranes is considered as a diffusion process. This is confirmed experimentally by measuring the noise spectra associated with pores of known dimension. It is then shown that one dimensional diffusion through pores of variable length can produce approximate 1/f noise spectra, if the distribution of lengths is proportional to (length)^–1.     

Surface potentials near the mouth of the large-conductance K+ channel from Chara australis: A new method of testing for diffusion-limited ion flow

The kinetics of single K⁺ channels were derived for patch-clamp recordings of membrane patches excised from cytoplasmic drops from the plant, Chara australis R. Br. Specifically, the tilt effect model of MacKinnon, Latorre and Miller (1989. Biochemistry 28:8092–8099) has been used to measure the electrostatic potential (surface PD) and fixed charge at the entrances of the channel. The surface PD is derived from the difference between the trans-pore potential difference (PD) and that between the two bulk phases. The trans-pore PD is probed using three voltage-dependent properties of the channel. These are (1) the association and dissociation rates of Ca²⁺ binding to the channel, from both the cytoplasmic and vacuolar solutions. These were determined from the mean blocked and unblocked durations of the channel in the presence of either 20 mmol liter^–1 vacuolar or 1 mmol liter^–1 cytoplasmic Ca²⁺; (2) the closing rate of the channel's intrinsic gating process. This was determined from the mean channel open time in the absence of vacuolar Ca²⁺ at membrane PDs more negative than –100 mV; and (3) the effect of Mg²⁺ on channel conductance when added to solutions initially containing 3 mmol liter^–1 KCl.The voltage dependence of properties 1 and 2 shifts along the voltage axis according to the ionic strength of the bathing media, consistent with the presence of negative charge in the channel vestibules. Furthermore, the magnitude of this shift depends on the current in a manner consistent with diffusion-limited ion flow in the channel (i.e., the rate of ion diffusion in the external electrolyte limits the channel conductance). Mg²⁺ on either side of the membrane alters channel conductance in a voltage-dependent way. A novel feature of the Mg²⁺ effect is that it reverses, from a block to an enhancement, when the membrane PD is more negative than –70 mV. This reversal only appears in solutions of low ionic strength. The attenuating effect is due to voltage-dependent binding of Mg²⁺ within the pore, which presumably plugs the channel. The enhancing effect is due to screening by Mg²⁺ of surface potentials arising from diffusion-limited flow of K⁺.     

Probing Pore Constriction in a Ligand-gated Ion Channel by Trapping a Metal Ion in the Pore upon Agonist Dissociation

Eukaryotic pentameric ligand-gated ion channels (pLGICs) are receptors activated by neurotransmitters to rapidly transport ions across cell membranes, down their electrochemical gradients. Recent crystal structures of two prokaryotic pLGICs were interpreted to imply that the extracellular side of the transmembrane pore constricts to close the channel (Hilf, R. J., and Dutzler, R. (2009) Nature 457, 115–118; Bocquet, N., Nury, H., Baaden, M., Le Poupon, C., Changeux, J. P., Delarue, M., and Corringer, P. J. (2009) Nature 457, 111–114). Here, we utilized a eukaryotic acetylcholine (ACh)-serotonin chimeric pLGIC that was engineered with histidines to coordinate a metal ion within the channel pore, at its cytoplasmic side. In a previous study, the access of Zn²⁺ ions to the engineered histidines had been explored when the channel was either at rest (closed) or active (open) (Paas, Y., Gibor, G., Grailhe, R., Savatier-Duclert, N., Dufresne, V., Sunesen, M., de Carvalho, L. P., Changeux, J. P., and Attali, B. (2005) Proc. Natl. Acad. Sci. U.S.A. 102, 15877–15882). In this study, the interactions of Zn²⁺ with the pore were probed upon agonist (ACh) dissociation that triggers the transition of the receptor from the active conformation to the resting conformation (i.e. during deactivation). Application of Zn²⁺ onto ACh-bound open receptors obstructed their pore and prevented ionic flow. Removing ACh from its extracellular binding sites to trigger deactivation while Zn²⁺ is still bound led to tight trapping of Zn²⁺ within the pore. Together with single-channel recordings, made to explore single pore-blocking events, we show that dissociation of ACh causes the gate to shut on a Zn²⁺ ion that effectively acts as a “foot in the door.” We infer that, upon deactivation, the cytoplasmic side of the pore of the ACh-serotonin receptor chimera constricts to close the channel.     

Periaxonal surface calcium binding and distribution of charge on the faces of squid axon potassium channel molecules

Summary The calcium binding constant associated with external surface charge in a position to influence the voltage sensing charges for potassium channel gating appears to be 30 molar^–1, a value much larger than previously thought and in approximate agreement with that found for artificial membranes composed of the lipid brain phosphatidylserene. Fixed charge on the periaxonal membrane surface is distributed in such a way that much larger charges occur at a distance of at least 8 angstroms from the channel pore openings. The separation between the ion pathway and the channel gating charge appears to be greater than or equal to 8 angstroms. Periaxonal surface charge which is in a position to determine the surface potential for gating has a magnitude greater than or equal to one (negative) electronic charge per 182 square angstrom before calcium binding, which is reduced to –e/625 Å in a normal divalent ionic environment. With the normal divalent ionic composition of seawater the surface potential at a position to influence the gating voltage sensor is –15 millivolts relative to the bulk external potential. The external surface potential is –3 mV at the pore mouth. There appears to be a negligible amount of fixed charge on the axoplasmic surface in the vicinity of the ion channel opening. Further, our results confirm earlier measurements that have given a negligible amount of axoplasmic surface fixed charge whose field components would be in a position to influence the channel gating charges.     

1/f Membrane noise generated by diffusion processes in unstirred solution layers

A mathematical treatment is given for 1/f noise observed in the ion transport through membranes. It is shown that this noise can be generated by current or voltage fluctuations which occur after step changes of the membrane permeability. Due to diffusion polarization in the unstirred solution layers near the membrane these fluctuations exhibit a 1/t time course which produces noise with a 1/f frequency dependence. The spectral density of 1/f noise is calculated for porous membranes with random switches between a finite and zero pore permeability. A wide frequency range and a magnitude of 1/f noise are obtained which are compatible with experimental data of 1/f noise reported for nerve membranes.Supported by Deutsche Forschungsgemeinschaft, Sonderforschungsbereich 38 Membranforschung.     

Transinhibition and Voltage-gating in a Fungal Nitrate Transporter

We have applied enzyme kinetic analysis to electrophysiological steady-state data of Zhou et al. (Zhou, J.J., Trueman, L.J., Boorer, K.J., Theodoulou, F.L., Forde, B.G., Miller, A.J. 2000. A high-affinity fungal nitrate carrier with two transport mechanisms. J. Biol. Chem. 275:39894–9) and to new current-voltage-time records from Xenopus oocytes with functionally expressed NrtA (crnA) 2H⁺-NO ₃ ^– symporter from Emericella (Aspergillus) nidulans. Zhou et al. stressed two Michaelis-Menten (MM) mechanisms to mediate the observed nitrate-induced currents, I _{NO 3} ^– . We show that a single straightforward reaction cycle describes the data well, pointing out that during exposure to external substrate, S = (2H⁺+NO ₃ ^– )_o, the product concentration inside, [P] = [H⁺] _i ² · [NO ₃ ^– _i, may rise substantially near the plasma membrane, violating the condition [P] [S] for MM kinetics. Here, [P] and its changes during experimentation are treated explicitly. K _1/2 20 µM for I _{NO 3} ^– at pH_o from Zhou et al. is confirmed. According to our analysis, NrtA operates between about 0.2 and 0.6 of the electrical distance in the membrane (outside 0, inside 1). In absence of thermodynamic gradients, the predominant orientation of the binding site(s) is probably inwards. The activity of the enzyme is sensitive to the transmembrane voltage, V, with an apparent gating charge of +1.0 ± 0.5 for inactivation, and transition probabilities of 0.3–1.3 s^–1 at V = 0. This gating mode impedes loss of cellular NO ₃ ^– during depolarization.     

1/f noise in black lipid membranes induced by ionic channels formed by chemically dimerized gramicidin A

Summary The noise behavior of lipid bilayer membranes, doped with a chemically dimerized gramicidin A, was investigated. In contrast to normal gramicidin A, which generates a Lorentzian type power spectrum due to the formation and disappearance of conducting dimers, the current power spectrum densityS _m (f) obtained with this gramicidin A derivative showed over several orders of magnitude a clear 1/f behavior. The intensity of this 1/f component was analyzed as a function of the membrane-applied voltage, membrane resistance, electrolyte concentration, and composition. The relationship between the meansquare fluctuation in current and the membrane current mean value was found to follow Hooge's equation, i.e., I ²=I _m ² /N f whereN is the number of channels and is a constant equal to 1.0×10^–2. It is suggested that a 1/f type noise was observed because the chemically dimerized form of gramicidin A produces long lasting cation selective channels.     

Further evidence for the regulation of the tight junction ion selectivity by cAMP in goldfish intestinal mucosa

Summary It has been reported that cAMP controls the transepithelial Cl^– conductance in fish intestine (Bakker, R., Groot, J.A., 1984,Am. J. Physiol. 246:G213–G217; Krasny, E.J., Madara, J.L., DiBona, D.L., Frizzell, R.A., 1983,Fed. Proc. 42:1100). In both studies, the cAMP effect was interpreted as an increase in tight junction Cl^– conductance, because cAMP did not change the membrane potential or membrane resistance ratio. However, the activation of a Cl^– conductance in the membranes of a subset of the epithelial cells might be difficult to discern from an increase in tight junction Cl^– conductance. Here we report experiments that were designed to distinguish a tight junction Cl^– conductance from a membrane Cl^– conductance in a subpopulation of the epithelial cells. The effect of hypotonicity on the cAMP-induced increase in transepithelial conductance showed that cAMP-induced conductance is located in series with the lateral intercellular spaces. Transepithelial serosa to mucosa direct current caused an increase in resistance due to so-called transport number effects. Forskolin abolished the transport number effects, indicating that cAMP increases the Cl^– conductance of the tight junctions. Increasing cAMP did not change mannitol fluxes, whereas Cl^– fluxes more than doubled. Changes in dilution potential and transepithelial resistance demonstrated that the cAMP-induced conductance is specific for Cl^– and Br^– as opposed to I^–, NO ₃ ^– , SO ₄ ^2– and gluconate^–. In contranst, cytochalasin D also decreased the transepithelial resistance and dilution potential in Nagluconate Ringer's. This demonstrates that cAMP acts on the tight junctions in a more specific manner than cytochalasin D.     

Phosphorylation of a peptide related to subunit c of the F0F1-ATPase/ATP synthase and relationship to permeability transition pore opening in mitochondria

A phosphorylated polypeptide (ScIRP) from the inner membrane of rat liver mitochondria with an apparent molecular mass of 3.5 kDa was found to be immunoreactive with specific antibodies against subunit c of F₀F₁-ATPase/ATP synthase (Azarashvily, T. S., Tyynelä, J., Baumann, M., Evtodienko, Yu. V., and Saris, N.-E. L. (2000). Biochem. Biophys. Res. Commun. 270, 741–744. In the present paper we show that the dephosphorylation of ScIRP was promoted by the Ca²⁺-induced mitochondrial permeability transition (MPT) and prevented by cyclosporin A. Preincubation of ScIRP isolated in its dephosphorylated form with the mitochondrial suspension decreased the membrane potential (_M) and the Ca²⁺-uptake capacity by promoting MPT. Incorporation of ScIRP into black-lipid membranes increased the membrane conductivity by inducing channel formation that was also suppressed by antibodies to subunit c. These data indicate that the phosphorylation level of ScIRP is influenced by the MPT pore state, presumably by stimulation of calcineurin phosphatase by the Ca²⁺ used to induce MPT. The possibility of ScIRP being part of the MPT pore assembly is discussed in view of its capability to induced channel activity.     

Reversible electrical breakdown of lipid bilayer membranes: A charge-pulse relaxation study

Summary Charge-pulse experiments were performed with lipid bilayer membranes from oxidized cholesterol/n-decane at relatively high voltages (several hundred mV). The membranes show an irreversible mechanical rupture if the membrane is charged to voltages on the order of 300 mV. In the case of the mechanical rupture, the voltage across the membrane needs about 50–200 sec to decay completely to zero. At much higher voltages, applied to the membrane by charge pulses of about 500 nsec duration, a decrease of the specific resistance of the membranes by nine orders of magnitude is observed (from 10⁸ to 0.1 cm²), which is correlated with the reversible electrical breakdown of the lipid bilayer membrane. Due to the high conductance increase (breakdown) of the bilayer it is not possible to charge the membrane to a larger value than the critical potential differenceV _c. For 1m alkali ion chloridesV _c was about 1 V. The temperature dependence of the electrical breakdown voltageV _c is comparable to that being observed with cell membranes.V _c decreases between 2 and 48°C from 1.5 to 0.6 V in the presence of 1m KCl.Breakdown experiments were also performed with lipid bilayer membranes composed of other lipids. The fast decay of the voltage (current) in the 100-nsec range after application of a charge pulse was very similar in these experiments compared with experiments with membranes made from oxidized cholesterol. However, the membranes made from other lipids show a mechanical breakdown after the electrical breakdown, whereas with one single membrane from oxidized cholesterol more than twenty reproducible breakdown experiments could be repeated without a visible disturbance of the membrane stability.The reversible electrical breakdown of the membrane is discussed in terms of both compression of the membrane (electromechanical model) and ion movement through the membrane induced by high electric field strength (Born energy).     

Calcium ions in the presence of exogenous phosphatidylserine interfere with GABA diffusion through the Deiters' neuron membrane

Diffusion of GABA through the plasma membrane of GABA-acceptive neurons might be a mechanism of importance for the termination of its synaptic action. In the present investigation we studied the effects of phosphatidylserine (PS) (10^–4–10^–3 M), Ca²⁺ 2 mM and PS+2 mM Ca²⁺ on such a process. The method involved the use of single microdissected Deiters' membranes which were put between two small microchambers in order to study the passage of GABA across the membrane. The results show that whereas PS and Ca²⁺ by themselves have no effect on such a process, PS+2 mM CaCl₂ give a significant, although slight, inhibition. The hypothesis that Calcium ion + PS effect is due to a disturbance of the interaction between GABA and endogenous PS molecules of the membrane is discussed.     

Determination of ion permeability through the channels made of porins from the outer membrane ofSalmonella typhimurium in lipid bilayer membranes

Summary The three types of porin (matrix-proteins) fromSalmonella typhimurium with molecular weights of 38,000, 39,000 and 40,000 were reconstituted with lipid bilayer membranes either as a trimer or as an oligomer (complex I). The specific conductance of the membranes increased several orders of magnitude after the addition of the porins into the aqueous phase bathing the membranes. A linear relationship between protein concentration in the aqueous phase and membrane conductance was found. In the case of lower protein concentrations (10^–12 m), the conductance increased in a stepwise fashion with a single conductance increment of 2.3 nS in 1m KCl. For a given salt the conductance increment was found to be largely independent of the particular porin (38 K, 39K or 40 K) and on the state of aggregation, although porin oligomers showed an up to 10 times smaller conductance increase in macroscopic conductance measurements. The conductance pathway has an ohmic current voltage characteristic and a poor selectivity for different alkali ions. Further information on the structure of the pores formed by the different porins fromSalmonella was obtained from the selectivity for various ions. From the permeability of the pore for large ions (Tris⁺, glucosamine⁺, Hepes^–_ a minimum pore diameter of 0.8 nm is estimated. This value is in agreement with the size of the pore as calculated from the conductance data for 1m KCl (1.4 nm for a pore length of 7.5 nm). The pore diameter may well account for the sugar permeability which has been found in reconstituted vesicles. The findings reported here are consistent with the assumption that the different porins form large aqueous channels in the lipid bilayer membranes and that the single condutance unit is a trimer. In addition, it is suggested that one trimer contains only one pore rather than a bundle of pores.     

Channel formation kinetics of gramicidin A in lipid bilayer membranes

Summary Previous studies have given evidence that the active form of gramicidin A in lipid bilayer membranes is a dimer which acts as an ion channel; it has been further shown that the mean lifetime of the channel strongly depends on the membrane thickness. As the thickness slightly decreases when a voltage is applied to the membrane, the equilibrium between conducting dimers and nonconducting monomers may be displaced by a voltage jump. From the relaxation of the electrical current after the voltage jump, information about the kinetics of channel formation is obtained. For a dioleoyllecithin/n-decane membrane the rate constant of association is found to be 2×10¹⁴ cm² mole^–1 sec^–1, which is by three orders of magnitude below the limiting value of a diffusion-controlled reaction in a two-dimensional system. The dissociation rate constant is equal to 2 sec^–1, a value which is consistent with the channel lifetime as obtained from electrical fluctuation measurements.     

Calcium-promoted resonance energy transfer between fluorescently labeled proteins during aggregation of chromaffin granule membranes

Proteins of the chromaffin granule membrane were covalently labeled in situ with sulfhydryl-specific fluorophores. Using MIANS (maleimide iodoaminonaphthyl sulfonate) as the donor and fluorescein mercury acetate or fluorescein-5-maleimide as the acceptor, Förster fluorescence resonance energy transfer (FRET) could be employed to measure the degree of inter-membrane and intra-membrane protein-protein contact upon Ca²⁺-induced aggregation of the membranes. The four major findings were: (1) Raising the Ca²⁺ concentration to approx. 500 μM causes the proteins to aggregate in the plane of the membrane. This is demonstrated by Ca²⁺-induced increases in the fluorescence resonance energy transfer in double labeled membranes. This effect is not protein-concentration dependent and occurs at calcium concentrations too low for granule aggregation, implying intra-membrane protein clustering or patching. To our knowledge this is the first direct demonstration of the fluid mosaic nature of subcellular organelles. (2) If two sets of granules are labeled separately, Ca²⁺-induced aggregation brings at least 74% of the labeled proteins into close transmembrane proximity. This effect is also observed at 10–100-fold slower rates in the absence of calcium and can be greatly reduced by depleting the granule membrane of labeled peripheral proteins. It is enhanced if the granules are aggregated by Ca²⁺ or K⁺. We conclude that (some) peripheral proteins can transfer from one membrane surface to another. (3) Aggregation of separately labeled sets of membranes by Ca²⁺ also produces transmembrane energy transfer since: (a) the K_m for Ca²⁺-induced quantum transfer is in the same range as the K_m for aggregation; (b) the reaction is protein-concentration dependent; (c) reversal of aggregation also (partially) reverses donor quenching. (4) A kinetic analysis of the transmembrane effect shows it to be 5–10-fold slower than aggregation itself, supporting earlier suggestions (Haynes, D.H., Kolber, M. and Morris, S.J., (1979) J. Theor. Biol. 81, 713–743) that lipid and protein rearrangements are secondary to granule membrane aggregation.     

Mechanism of ion transport through lipid bilayer-membranes mediated by peptide cyclo-(d-Val-l-Pro-l-Val-d-Pro)3

The cyclic dodecapeptide PV, cyclo-(d-Val-l-Pro-l-Val-d-Pro)₃, a structural analogue of the ion-carier valinomycin, increase the cation permeability of lipid bilayer membranes. This paper reports the results of two types of relaxation experiments, namely relaxation of the membrane current after a voltage jump and decay of the membrane voltage after a charge pulse in lipid bilayer membranes exposed to PV. From the relaxation data, the rate constant for the translocation of the ion carrier complex across the membrane, as well as the partition coefficient of the complex between water and membrane solution interface were computed and found to be about one order of magnitude less than the comparable values for valinomycin (Val). Furthermore, the dependence of the initial membrane conductivity on ion concentration was used to evaluate the equilibrium constant, K, of complexation between PV and some monovalent cations in water. The values of K yield the following selectivity sequence of PV: Na⁺ < NH₄⁺ < K⁺ < Cs⁺ < Rb⁺. These and earlier results are consistent with the idea that PV promotes cation movement across membranes by the solution complexation mechanism which involves complexation between ion and carrier in the aqueous phase and transport of the carrier across the membrane. In the particular form of the solution complexation mechanism operating here, the PV present in the PV-cation complex carrying charge across the membrane derives from the side from which the current is flowing (cis-mechanism). As shown previously, valinomycin, in contrast to PV, acts by an interfacial complexation mechanism in which the Val in the Val-cation complex derives from the side toward which current is flowing (trans-mechanims). The comparison of the kinetic properties of these two closely related compounds yields interesting insights into the relationship between chemical structure and function of ion carriers.     

Noise analysis and relaxation experiments of transport of hydrophobic anions across lipid membranes at equilibrium and nonequilibrium

Under equilibrium and nonequilibnum steady-stale conditions the spectral intensity of current noise S_J(f) generated by the transport of hydrophobic unions across lipid bilayer membranes was investigated. The experimental results were compared with different reaction models S_J(f) showed a characteristic increase proportional to f² between frequency-independent tails at low and high frequencies. This gradient was found to be independent of applied voltage which indicates the contribution of a single voltage-dependent reaction step of ion translocation across the membrane From the shape of S_J(f) at low frequencies the rate constant of ion desorption from the membrane into the aqueous phase could be estimated. Unambiguous evidence for the application of a general model, which includes the coupling of slow ion diffusion in the aqueous phase to ion adsorption/desorption at the membrane interface, could not be obtained from the low-frequency shape of S_J(f). The shot noise of this ion transport determines the amplitude of S_J(f) at high frequencies which decreases with increasing voltage applied. Analysis of voltage-jump current-relaxation experiments and of current noise carried cut on one membrane yielded significant differences of the derived ion partition coefficient. This deviation is qualitatively described on the basis of incomplete reaction steps.     

Band-selective 3D NOESY-TOCSY: Measurement of through-space correlations between aliphatic protons of membrane peptides and proteins in non-deuterated detergents

Summary Recently the use of band-selective excitation to obtain ¹H 2D NMR spectra of membrane peptides and proteins in non-deuterated detergents has been demonstrated [Seigneuret, M. and Levy, D. (1995) J. Biomol. NMR, 5, 345–352]. A limitation of the method was the inability to obtain through-space correlation between aliphatic protons. Here, a 3D F3-band-selective NOESY-TOCSY experiment is described that allows such correlations to be observed in the presence of an excess of non-deuterated detergent. Application to the measurement of proximities between aliphatic protons of the membrane peptide mastoparan X solubilized in non-deuterated n-octylglucoside is presented. With this additional experiment, it is now possible to obtain the same amount of structural constraints on membrane peptides and protein in non-deuterated detergent as in deuterated detergent and therefore to perform complete structural studies.     

Electrolyte permeability and electromotive action of mosaic membranes composed of porous and liquid parts of high ionic selectivity for ions of opposite signs

Summary A theory is presented of the electromotive and ion permeability properties of membranes which consist of a mosaic of highly ion selectiveporous membrane areas of ion exchanger nature and of areas of highly ion selectiveliquid ion exchanger membranes, the two types of areas being exclusively permeable to ions of opposite sign. It is demonstrated that, with properly chosen membranes, the preferential permeability of such porous-liquid mosaic membranes for ions of one sign of charge will be the opposite of that apparently indicated by their electromotive action.The theory is based on the fact that the movement of ions across the porous membranes occurs in the dissociated state and in most instances is quantitatively linked to the resistance according to the Nernst-Einstein equation. The penetration of ions across liquid ion exchanger membranes, however, takes place essentially in a nondissociated state, and, as determined by self-exchange studies with radioactive tracers and stirred membranes, occurs at rates far in excess of those across porous membranes of the same resistance.For the theoretical treatment the simplest case, two-membrane macro-model concentration cells, is discussed in detail. Qualitatively, it is evident that the ratio of the permeability of anions and cations across such porous-liquid mosaic membranes ordinarily will be strongly in favor of the ions which penetrate across its liquid parts; contrariwise, the electromotive actions of the mosaic membranes ordinarily are dominated by its porous parts.Electric currents flow through all mosaic membranes; the strenghth of the current in a model cell may be calculated from the concentration potentials arising separately across the two membranes, and the resistances of the membranes and of the two solutions. From the strength of the current, the sign and the magnitude of the concentration potential arising in the model cell may be computed; in many instances it should closely approach the concentration potential across the porous membrane.For the test of this theory with two-membrane macro-mosaic models, the electrolyte of choice for experimental reasons was RbSCN, tagged with⁸⁶Rb⁺ and S¹⁴CN^–. The porous membranes were polystyrene sulfonic acid collodion matrix membranes; the liquid membranes consisted of 0.02m trioctyl propyl ammonium thiocyanate in 1-decanol. The ratios of the permeabilities across the model mosaic membranes determined by conventional rate of self-exchange measurements showed, as expected, that the permeability of the SCN^– ions is larger, up to 3600 times larger, than that of the Rb⁺ ions. The potentials arising in these models agreed within the limits of experimental error with those predicted by theory, closely approaching that arising at the cation selective porous membranes.