Coupled transepithelial sodium and potassium transport across isolated frog skin: effect of ouabain, amiloride and the polyene antibiotic filipin

Pump current and Na+/K+ coupling ratio of Na+/K+-ATPase in reconstituted lipid vesicles

A method is described for studying the coupling ratio of the Na+/K+ pump, i.e., the ratio of pump-mediated fluxes of Na+ and K+, in a reconstituted system. The method is based on the comparison of the pump-generated current with the rate of K+ transport. Na+/K+-ATPase from kidney is incorporated into the membrane of artificial lipid vesicles; ATPase molecules with outward-oriented ATP-binding site are activated by addition of ATP to the medium. Using oxonol VI as a potential-sensitive dye for measuring transmembrane voltage, the pump current is determined from the change of voltage with time t. In a second set of experiments, the membrane is made selectively K+-permeable by addition of valinomycin, so that the membrane voltage U is equal to the Nernst potential of K+. Under this condition, dU/dt reflects the change of intravesicular K+ concentration and thus the flux of K+. Values of the Na+/K+ coupling ratio determined in this way are close to 1.5 in the experimental range (10-75 mM) of extravesicular (cytoplasmic) Na+ concentrations.

Calcium modulates transepithelial potassium secretion across isolated frog skin

1. Transepithelial K+ movements across isolated frog skin consist of four components: (i) a passive component; (ii) an active inward transport of K+ which occurs via the epithelial cells; (iii + iv) two active outward-directed components, one via the skin glands, the other via the epithelial cells. 2. Incubation of frog skin in gluconate Ringer's solution activates the K+ secretion via the epithelial cells. 3. A decrease in the Ca2+ activity of the epithelial cells increases the K+ permeability of the apical membrane and reduces the K+ permeability of the basolateral membrane.

Interactions of sodium transport, cell volume, and calcium in frog urinary bladder

The volume of individual cells in intact frog urinary bladders was determined by quantitative microscopy and changes in volume were used to monitor the movement of solute across the basolateral membrane. When exposed to a serosal hyposmotic solution, the cells swell as expected for an osmometer, but then regulate their volume back to near control in a process that involves the loss of KCl. We show here that volume regulation is abolished by Ba++, which suggests that KCl movements are mediated by conductive channels for both ions. Volume regulation is also inhibited by removing Ca++ from the serosal perfusate, which suggests that the channels are activated by this cation. Previously, amiloride was observed to inhibit volume regulation: in this study, amiloride-inhibited, hyposmotically swollen cells lost volume when the Ca++ ionophore A23187 was added to Ca++-replete media. We attempted to effect volume changes under isosmotic conditions by suddenly inhibiting Na+ entry across the apical membrane with amiloride, or Na+ exit across the basolateral membrane with ouabain. Neither of these Na+ transport inhibitors produced the expected results. Amiloride, instead of causing a decrease in cell volume, had no effect, and ouabain, instead of causing cell swelling, caused cell shrinkage. However, increasing cell Ca++ with A23187, in both the absence and presence of amiloride, caused cells to lose volume, and Ca++-free Ringer's solution (serosal perfusate only) caused ouabain-blocked cells to swell. Finally, again under isosmotic conditions, removal of Na+ from the serosal perfusate caused a loss of volume from cells exposed to amiloride. These results strongly suggest that intracellular Ca++ mediates cell volume regulation by exerting a negative control on apical membrane Na+ permeability and a positive control on basolateral membrane K+ permeability. They also are compatible with the existence of a basolateral Na+/Ca++ exchanger.

Interaction between the basolateral K+ and apical Na+ conductances in Necturus urinary bladder

Experimental modulation of the apical membrane Na+ conductance or basolateral membrane Na+-K+ pump activity has been shown to result in parallel changes in the basolateral K+ conductance in a number of epithelia. To determine whether modulation of the basolateral K+ conductance would result in parallel changes in apical Na+ conductance and basolateral pump activity, Necturus urinary bladders stripped of serosal muscle and connective tissue were impaled through their basolateral membranes with microelectrodes in experiments that allowed rapid serosal solution changes. Exposure of the basolateral membrane to the K+ channel blockers Ba2+ (0.5 mM/liter), Cs+ (10 mM/liter), or Rb+ (10 mM/liter) increased the basolateral resistance (Rb) by greater than 75% in each case. The increases in Rb were accompanied simultaneously by significant increases in apical resistance (Ra) of greater than 20% and decreases in transepithelial Na+ transport. The increases in Ra, measured as slope resistances, cannot be attributed to nonlinearity of the I-V relationship of the apical membrane, since the measured cell membrane potentials with the K+ channel blockers present were not significantly different from those resulting from increasing serosal K+, a maneuver that did not affect Ra. Thus, blocking the K+ conductance causes a reduction in net Na+ transport by reducing K+ exit from the cell and simultaneously reducing Na+ entry into the cell. Close correlations between the calculated short-circuit current and the apical and basolateral conductances were preserved after the basolateral K+ conductance pathways had been blocked. Thus, the interaction between the basolateral and apical conductances revealed by blocking the basolateral K+ channels is part of a network of feedback relationships that normally serves to maintain cellular homeostasis during changes in the rate of transepithelial Na+ transport.

Relationships among sodium current, permeability, and Na activities in control and glucocorticoid-stimulated rabbit descending colon

Effects of a potent synthetic glucocorticoid, methylprednisolone (MP), on transepithelial Na transport were examined in rabbit descending colon. Current-voltage (I-V) relations of the amiloride-sensitive apical Na entry pathway were measured in colonic tissues of control and MP-treated (40 mg im for 2 days) animals. Tissues were bathed mucosally by solutions of various Na activities, (Na)m, ranging from 6.2 to 75.6 mM, and serosally by a high K solution. These I-V relations conformed to the "constant field" flux equation permitting determination of the permeability of the apical membrane to Na, PmNa, and the intracellular Na activity, (Na)c. The following empirical relations were observed for both control and MP-treated tissues: Na transport increases hyperbolically with increasing (Na)m obeying simple Michaelis-Mentin kinetics; PmNa decreased hyperbolically with increasing (Na)m, but was unrelated to individual variations in (Na)c; (Na)c increased hyperbolically with (Na)m; both spontaneous and steroid-stimulated variations in Na entry rate could be attributed entirely to parallel variations in PmNa at each mucosal Na activity. Comparison of these empirical, kinetic relations between control and MP-treated tissues revealed: maximal Na current and PmNa were greater in MP tissues, but the (Na)m's at which current and PmNa were half-maximal were markedly reduced; (Na)c was significantly increased in MP tissues at each (Na)m while the (Na)m at half-maximal (Na)c was unchanged. These results provide direct evidence that glucocorticoids cause marked stimulation of Na absorption across rabbit colon primarily by increasing the Na permeability of the apical membrane. While the mechanism for the increased permeability remains to be determined, the altered relation between PmNa and (Na)m suggests possible differences in the conformation or environment of the Na channel in MP-treated tissues.

Ba2+-induced changes in the Na+- and K+-permeability of the isolated frog skin

Addition of the K+-channel blocking agent Ba2+ to the basolateral solution (in a concentration which is assumed to block the K+-flux via the K+-channels completely) resulted initially in a two-thirds reduction in the short-circuit current (SCC), followed by a complete recovery of the SCC. To examine the reason for this recovery, experiments were carried out which made it possible to calculate the Na+-permeability of the apical membrane (PaNa) and the K+-permeability of the basolateral membrane (PbK). The presence of Ba2+ had no significant effect on the cell volume and the cellular Na+- and K+-concentration. Addition of Ba2+ resulted in a depolarization of the intracellular potential (VSCC) from a control value of -76.3 +/- 2.8 mV to -15.1 +/- 1.7 mV. Although a complete recovery in the SCC was observed, VSCC did not recover. The K+-flux across the basolateral membrane was estimated from washout experiments. The washout of 42K+ (the K+-efflux) could be described by a single exponential component with a half time of 30-70 min. The addition of Ba2+ during the washout resulted in a transient decrease in 42K+-efflux from the epithelium. From VSCC and the cellular K+ and Na+-concentration and the coupling ratio of the Na-K pump, it was found that Na+-permeability of the apical membrane was 6.5 X 10(-7) cm X s-1 before the addition of Ba2+ and 1.7 X 10(-6) cm X s-1 when the SCC had recovered after the addition of Ba2+ and PbK changed from 8.8 X 10(-6) cm X s-1 to 1.5 X 10(-6) cm X s-1. Thus, the observed recovery in SCC was due to a considerable increase in Na+-permeability of the apical membrane and the presence or appearance of a small Ba2+-insensitive K+-permeability in the basolateral membrane.

Effects of anions on cellular volume and transepithelial Na+ transport across toad urinary bladder

The effects of complete substitution of gluconate for mucosal and/or serosal medium Cl- on transepithelial Na+ transport have been studied using toad urinary bladder. With mucosal gluconate, transepithelial potential difference (VT) decreased rapidly, transepithelial resistance (RT) increased, and calculated short-circuit current (Isc) decreased. Calculated ENa was unaffected, indicating that the inhibition of Na+ transport was a consequence of a decreased apical membrane Na+ conductance. This conclusion was supported by the finding that a higher amiloride concentration was required to inhibit the residual transport. With serosal gluconate VT decreased, RT increased and Isc fell to a new steady-state value following an initial and variable transient increase in transport. Epithelial cells were shrunken markedly as judged histologically. Calculated ENa fell substantially (from 130 to 68 mV on average). Ba2+ (3 mM) reduced calculated ENa in Cl- Ringer's but not in gluconate Ringer's. With replacement of serosal Cl- by acetate, transepithelial transport was stimulated, the decrease in cellular volume was prevented and ENa did not fall. Replacement of serosal isosmotic Cl- medium by a hypo-osmotic gluconate medium (one-half normal) also prevented cell shrinkage and did not result in inhibition of Na+ transport. Thus the inhibition of Na+ transport can be correlated with changes in cell volume rather than with the change in Cl-per se. Nystatin virtually abolished the resistance of the apical plasma membrane as judged by measurement of tissue capacitance. With K+ gluconate mucosa, Na+ gluconate serosa, calculated basolateral membrane resistance was much greater, estimated basolateral emf was much lower, and the Na+/K+ basolateral permeability ratio was much higher than with acetate media. It is concluded the decrease in cellular volume associated with substitution of serosal gluconate for Cl results in a loss of highly specific Ba2+-sensitive K+ conductance channels from the basolateral plasma membrane. It is possible that the number of Na+ pump sites in this membrane is also decreased.

Current-voltage relations of the basolateral membrane in tight amphibian epithelia: use of nystatin to depolarize the apical membrane

Exposure of the mucosal side of toad (Bufo bufo) urinary bladder and frog (Rana ridibunda) skin to the polyene ionophore nystatin, resulted in stable preparations in which the apical resistance was negligible compared to the basolateral resistance. The preparations support passive K currents in both directions and an amiloride-insensitive Na current in the apical-serosal direction which is blocked by ouabain. The nystatin-treated toad bladder was used to study the electrical properties of the basolateral membrane by means of current-voltage curves recorded transepithelially. The K current showed strong rectification at cellular potentials negative with respect to the interstitial space. The ouabain-sensitive current increased with membrane voltage at negative voltages but saturated above +20 mV.

Active transepithelial potassium transport in frog skin via specific potassium channels in the apical membrane

In frog skin bathed in Cl--Ringer's solution the short circuit current (SCC) is equal to the net Na+ flux. In the present study Na+ and K+ transport across frog skin have been investigated in skins bathed in a solution where all Cl- has been substituted by the impermeable anion gluconate. In this solution the net Na+ flux (9.22 +/- 0.72 nmole/cm2/min) was significantly higher than the SCC (7.61 +/- 0.63) nmole/cm2/min). Measurement of the transepithelial K+ influx and K+ efflux showed that the discrepancy between the net Na+ flux and the SCC was caused by an active outwards going transepithelial K+ transport. The K+ but not the Na+ transport could be blocked by adding the K+ channel blocking agent Ba++ to the apical solution. Thus, the K+ transport occurs via a K+ specific pathway in the apical membrane. Ouabain blocked both the Na+ and the K+ transport, whereas the presence of the Na+ channel blocking agent amiloride in the apical solution blocked the Na+ transport and reduced the K+ transport. In the presence of amiloride in the apical solution the SCC and the transepithelial potential difference (PD) reversed so that the outside (the apical side) of the frog skin became positive with respect to the basolateral side. The inverted SCC was carried by an active transepithelial K+ transport, this K+ transport required the presence of Na+ in the basolateral solution. The experiments show that frog skin can insert or activate K+ channels in the apical membrane, indicating that the frog may regulate its K+ content by varying the K+ permeability of the apical membrane.

Microelectrode study of K+ accumulation by tight epithelia: I. Baseline values of split frog skin and toad urinary bladder

Toad bladder and split frog skin were impaled with fine-tipped single- and double-barrelled K+-selective microelectrodes. In order to circumvent membrane damage induced by impaling toad bladder, a null point method was developed, involving elevations of mucosal potassium concentration. The results suggest that intracellular potassium activity of short-circuited toad bladder is approximately 82 mM, twice as large as earlier estimates. Far more stable and rigorously defined intracellular measurements were recorded from short-circuited split frog skins. The intracellular positions of the micropipette and microelectrode tips were verified by transient hyperpolarizations of the membrane potential with mucosal amiloride or by transient depolarizations with serosal barium or strophanthidin. Simultaneous impalement of distant cells with separate micropipettes demonstrated that both the baseline membrane potentials and the responses to depolarizing agents were similar, further documenting that frog skin is a functional syncytium. Measurements with double-barrelled microelectrodes and simultaneous single-barrelled microelectrodes and reference micropipettes suggest that the intracellular potassium activity is about 104 mM, lower than previously reported. Taken together with measurements of intracellular potassium concentration, this datum suggests that potassium is uniformly distributed within the epithelial cells.

Apical membrane permeability and kinetic properties of the sodium pump in rabbit urinary bladder

Previous studies have shown that aldosterone stimulates the rate of Na+ transport across the rabbit urinary bladder epithelium by increasing the apical membrane permeability to Na+. Paradoxically, ion-sensitive and conventional micro-electrode measurements demonstrated that intracellular Na+ activity aiNa+ was essentially unchanged by aldosterone, i.e. aiNa+ was constant regardless of the rate of Na+ transport. The present study was designed to resolve this apparent contradiction. The effects of elevated, endogenous aldosterone levels produced by low-Na+ diet (Lewis & Diamond, 1976) on urinary bladder Na+ transport were investigated in vitro using Ussing-type chambers and intracellular conventional and ion-sensitive microelectrodes. Apical membrane selectivity and kinetics of the Na+ pump were assessed as a function of hormone stimulation. The aldosterone-stimulated increase in Na+ transport was accounted for by increases in both the relative selective permeability of the apical membrane to Na+ and an increase in its absolute Na+ permeability. The kinetics of the Na+ pump were evaluated electrically by loading the cells with Na+ (monitored with Na+-sensitive micro-electrodes) or alternatively by manipulating serosal solution K+ concentration and measuring changes in the basolateral membrane electromotive forces and resistance. From these measurements the current generated by the pump was calculated as a function of intracellular Na+ or extracellular K+. The kinetics of the pump were not altered by aldosterone. A model of highly co-operative binding estimated Km for Na+ as 14.2 mM and 2.3 mM for K+. Hill coefficients for these ions were 2.8 and 1.8, respectively, consistent with a pump stoichiometry of 3 Na+ to 2 K+. The kinetic properties of the Na-K pump indicate that physiological levels of aiNa+ are poised at the foot of a step kinetic curve which energetically favours Na+ extrusion.

Epithelial transport parameters: an analysis of experimental strategies

A set of experiments was simulated on a computer version of the Koefoed-Johnsen & Ussing model for high-resistance epithelia. The results obtained were analysed according to procedures commonly applied to the analyses of experimental data and interpreted in terms of the model parameters. Although the computer model encodes a stoichiometry of 3:2 for Na-K exchange through the Na pump, the simulation of published experimental procedures yields different figures in almost every case. We show that ENa as originally defined by Ussing & Zerahn (Acta physiol. scand. 23, 110-127 (1951)) and as obtained from flux-ratio experiments has different values under different experimental conditions with unchanged system parameters and that it is distinct from ENa measured by other methods. We also show that unless the pump is saturated with internal Na an increase in the rate of pumping cannot cause a substantial increase in the rate of transepithelial Na transport.

Electrical properties of the rabbit urinary bladder assessed using gramicidin D

Recently, antibiotics have enjoyed widespread usage as tools in studies of epithelial transport. In the present study we assess the usefulness of the pore-forming antibiotic gramicidin D as a means for probing the electrical properties of the tight epithelium rabbit urinary bladder. Addition of 50 micron gramicidin to the mucosal bath (either a NaCl or KCl Ringer's solution) led to a large irreversible increase in the transepithelial conductance (GT) with 800 sec. GT increased by approximately 1200% and 500% in KCl and NaCl Ringer's solutions, respectively. Microelectrode measurements of the resistance ratio (the ratio of apical membrane resistance to basolateral membrane resistance) showed that apical membrane resistance is decreased by the drug. Measurements of the basolateral membrane resistance (Rbl) and tight junctional resistance (Rj) using a new and independent method (based on the perturbation of basolateral membrane electrogenic Na+ pump) demonstrated that Rbl and Rj were unaffected, suggesting that the effects of gramicidin are restricted to the apical membrane for periods of at least 2 hours after drug addition. The selectivity of the gramicidin-induced permeability in the apical membrane was calculated from measurements of the apical membrane potential after ion substitutions using a modified version of the constant field equation. The selectivity sequence for cations was Cs+ greater than K+ greater than Na+ greater than Li+ greater than choline. Unlike the commonly used polyene antibiotics nystatin and amphotericin B, gramicidin did not induce a significant Cl- permeability. In addition, the dose-response curve had a slope of 1. A method is described for calculating membrane resistances directly from transepithelial measurements under some conditions of gramicidin use, without requiring the use of microelectrode measurements.

The electrophysiology of rabbit descending colon. II. Current-voltage relations of the apical membrane, the basolateral membrane, and the parallel pathways

In this paper we employ the data described in the previous paper (I) to derive the current-voltage (I-V) relations of the basolateral membrane, the amiloride-insensitive "leak" pathway across the apical membrane, and the parallel pathways across rabbit descending colon. The results indicated that: a) The resistance of the basolateral membrane is independent of the electrical potential difference across that barrier over the range -8 to 67 mV and averaged 195 omega cm2. The electromotive force across this barrier averaged 50 mV under control conditions and 48 mV in the presence of amiloride. The origin of this difference is discussed. b) The resistance of the parallel pathways averaged 351 omegacm2 and was independent of the transepithelial electrical potential difference over the range -170 to + 90mV. The conductance of these pathways can be reasonably well accounted for by the partial ionic conductances of Na, K and Cl reported previously. c) The resistance of the amiloride-insensitive pathway across the apical membrane averaged 1667 omegacm2 and the electromotive force across this pathway averaged -51 mV. These values are in excellent agreement with those determined by others. The ionic nature of this "leak" pathway remains to be elucidated.

Potassium transport across the frog retinal pigment epithelium

Previous experiments indicate that the apical membrane of the frog retinal pigment epithelium contains electrogenic Na:K pumps. In the present experiments net potassium and rubidium transport across the epithelium was measured as a function of extracellular potassium (rubidium) concentration, [K]0 ( [Rb]0). The net rate of retina-to-choroid 42K(86Rb) transport increased monotonically as [K]0 ( [Rb]0) increased from approximately 0.2 to 5 mM on both sides of the tissue or on the apical (neural retinal) side of the tissue. No further increase was observed when [K]0 ( [Rb]0) was elevated to 10 mM. Net sodium transport was also stimulated by elevating [K]0. The net K transport was completely inhibited by 10-4 M ouabain in the solution bathing the apical membrane. Ouabain inhibited the unidirectional K flux in the direction of net flux but had no effect on the back-flux in the choroid-to-retina direction. The magnitude of the ouabain-inhibitable 42K(86Rb) flux increased with [K]0 ( [Rb]0). These results show that the apical membrane Na:K pumps play an important role in the net active transport of potassium (rubidium) across the epithelium. The [K]0 changes that modulate potassium transport coincide with the light-induced [K]0 changes that occur in the extracellular space separating the photoreceptors and the apical membrane of the pigment epithelium.

Mechanisms of K+ transport in isolated turtle urinary bladder. Induction of active K+ secretion in a K+-absorbing epithelium

Transepithelial K(+) movement was studied in vitro in the short-circuited turtle bladder by increasing luminal K(+) permeability and by inhibiting the basolateral Na/K pump. Luminal addition of amphotericin B caused net K(+) secretion (180+/-52 nmol/h) compared with net K(+) absorption (42+/-6 nmol/h) in control bladders. Serosal ouabain and luminal amiloride abolished K(+) secretion in amphotericin-treated bladders; ouabain restored net absorption (45+/-16 nmol/h). The direction and rate of net K(+) transport are controlled by the relative K(+) permeabilities and the Na/K pump sites at the two cell membranes of the epithelium.

Kinetic studies on the effects of ouabain on Na+ fluxes in frog skin

Among 48 pieces of paired frog skins of Rana pipiens in Ringer's solution, 10 pieces showed a strictly monotone decrease in the short circuit current (SCC) following ouabain treatment (10(-4) M). In 9 cases a transient attenuation, and in 27 cases a distinct wave in the ebb of the SCC, was seen. In 2 instances, two waves were seen. Associated with the not-monotone events was a transient rise in electrical skin conductance. The reasons for these mixed skin responses are unknown. One possible reason is considered here: Early during the ouabain action, some of the Na+ entering from the mucosal side is trapped in the skin by electroneutral processes, in keeping with the already known fact that ultimately cellular KCl is partly replaced by NaCl. Computer assisted model studies show how monotone, and not-monotone "transepithelial" net Na+ flux curves can be generated. Essential conditions for the generation of not-monotone Na+ flux curves are: 1. Presence of two distinct "cellular", active Na+ pools in the model. 2. Presence of a loop pathway in which a principal "transepithelial Na+ transport compartment", and a constituent "Na+/K+ maintenance compartment", are connected to each other and to the "extracellular" compartment. The model, then, predicts under which kinetic conditions monotone and not-monotone transepithelial Na+ flux curves will be seen.

Ouabain-sensitive 86Rb(K) influx is linked to transepithelial Na transport in pig kidney cell line

The pig kidney cell line, LLC-PK1, exhibits rheogenic D-glucose coupled transepithelial Na+ transport that is inhibited by phlorizin. By measuring the difference in initial rates of influx of 86Rb+ with and without coupled Na+ transport, we can demonstrate an 86Rb+ uptake linked to Na+ transport, The simultaneous determination of phlorizin-inhibited Na coupled D-[3H] glucose uptake and 86Rb+ influx allows calculation of an Na+/Rb+ stoichiometry that is consistent with an electrogenic Na+ for Rb+ exchange.

Electrophysiologic changes associated with potassium depletion of frog skin

Skins from the frog Rana pipiens pipiens were studied under short-circuited conditions during the course of removing and replacing potassium in the inner bathing media in 14 experiments. The intracellular potential (Vsc), fractional resistance (FR), short-circuit current (Isc) and total tissue conductance (gr) were constantly monitored during impalements of the epithelial cells. The mean value (+/- SE) for Vsc was --79 (+/- 3) mV under baseline conditions. Removal of potassium from the inner bathing solution transiently stimulated the short-circuit current and hyperpolarized the basolateral membrane; with sufficiently long incubations, the basolateral membrane was eventually depolarized. Restoration of potassium to the inner solution within 43 min after initiating the perfusion with K+-free solution depolarized the basolateral membrane. However, restoration of potassium after at least 1 1/2 hr of incubation hyperpolarized the membrane. Ouabain consistently depolarized the basolateral membrane, even after extended periods of potassium depletion as long as 320 min. In the presence of ouabain, restoration of potassium depolarized the basolateral membrane. The data provide further evidence for the concept that the Na--K exchange pump of frog skin is rheogenic. Furthermore, the results suggest that the pump continues to be active even during prolonged periods of potassium depletion, reaccumulating potassium which has leaked out of the epithelial cells.

Active sodium transport by turtle colon via an electrogenic Na-K exchange pump

Active sodium absorption by a variety of epithelia is abolished by ouabain, but the obligatory coupling between the movement of sodium and potassium expected from a basolateral (Na+ + K+) ATPase has not been convincingly demonstrated. According to the model of Koefoed-Johnsen and Ussing, the asymmetric cation selectivities of the apical and basolateral membranes prevent basolateral Na-K exchange from being expressed as opposing transmural ion flows. An additional consequence of this asymmetry is that the short-circuit current (Isc) cannot be identified with the current through the sodium-potassium pump. We used the polyene antibiotic, amphotericin-B, to reduce the resistance and the cation selectivity of the apical membrane of isolated turtle colon so that the basolateral membrane current could be dissected into two components: one through a barium-sensitive potassium channel and another which represents the current associated with ouabain-sensitive, electrogenic, Na-K exchange. Comparison of cation fluxes and short circuit current indicates that in these conditions active sodium absorption is entirely attributable to an electrogenic Na-K pump with a stoichiometry of approximately 3Na:2K.