[3H]bumetanide binding to avian erythrocyte membranes. Correlation with activation and deactivation of Na/K/2Cl cotransport

General models for water transport across leaky epithelia

The group of leaky epithelia, such as proximal tubule and small intestine, have several common properties in regard to salt and water transport. The fluid transport is isotonic, the transport rate increases in dilute solutions, and water can be transported uphill. Yet, it is difficult to find common features that could form the basis for a general transport model. The direction of transepithelial water transport does not correlate with the direction of the primary active Na+ transport, or with the ultrastucture as defined by the location of apical and basolateral membranes, of the junctional complex and the lateral intercellular spaces. The presence of specific water channels, aquaporins, increases the water permeability of the epithelial cell membranes, i.e., the kidney proximal tubule. Yet other leaky epithelia, for example, the retinal pigment epithelium, have no known aquaporins. There is, however, a general correlation between the direction of transepithelial transport and the direction of transport via cotransporters of the symport type. A simple epithelial model based on water permeabilities, a hyperosmolar compartment and restricted salt diffusion, is unable to explain epithelial transport phenomena, in particular the ability for uphill water transport. The inclusion of cotransporters as molecular water pumps in these models alleviates this problem.

Sodium-potassium-chloride cotransport

Obligatory, coupled cotransport of Na(+), K(+), and Cl(-) by cell membranes has been reported in nearly every animal cell type. This review examines the current status of our knowledge about this ion transport mechanism. Two isoforms of the Na(+)-K(+)-Cl(-) cotransporter (NKCC) protein (approximately 120-130 kDa, unglycosylated) are currently known. One isoform (NKCC2) has at least three alternatively spliced variants and is found exclusively in the kidney. The other (NKCC1) is found in nearly all cell types. The NKCC maintains intracellular Cl(-) concentration ([Cl(-)](i)) at levels above the predicted electrochemical equilibrium. The high [Cl(-)](i) is used by epithelial tissues to promote net salt transport and by neural cells to set synaptic potentials; its function in other cells is unknown. There is substantial evidence in some cells that the NKCC functions to offset osmotically induced cell shrinkage by mediating the net influx of osmotically active ions. Whether it serves to maintain cell volume under euvolemic conditons is less clear. The NKCC may play an important role in the cell cycle. Evidence that each cotransport cycle of the NKCC is electrically silent is discussed along with evidence for the electrically neutral stoichiometries of 1 Na(+):1 K(+):2 Cl- (for most cells) and 2 Na(+):1 K(+):3 Cl(-) (in squid axon). Evidence that the absolute dependence on ATP of the NKCC is the result of regulatory phosphorylation/dephosphorylation mechanisms is decribed. Interestingly, the presumed protein kinase(s) responsible has not been identified. An unusual form of NKCC regulation is by [Cl(-)](i). [Cl(-)](i) in the physiological range and above strongly inhibits the NKCC. This effect may be mediated by a decrease of protein phosphorylation. Although the NKCC has been studied for approximately 20 years, we are only beginning to frame the broad outlines of the structure, function, and regulation of this ubiquitous ion transport mechanism.

Na(+)-K(+)-2Cl- cotransport, Na+/H+ exchange, and cell volume in ferret erythrocytes

Ferrets have high-Na+ and low-K+ erythrocytes (113 and 5.4 mmol/l cell water) due to the lack of Na(+)-K+ pumps. Because ferret erythrocytes have a high capacity for Na(+)-K(+)-2Cl- cotransport, the present study was undertaken to evaluate cell volume-related changes in cotransport activity and its role in volume regulation. With cell shrinkage, Na(+)-K(+)-2Cl- cotransport is activated about twofold. A large bumetanide-insensitive Na+ uptake component that has not yet been described is found in shrunken erythrocytes. Its inhibition by amiloride (concn inhibiting 50% of maximal response = 12 microM) and the Na+ dependence of amiloride-sensitive extracellular pH changes measured in cells suspended in hypertonic unbuffered medium indicate that this flux represents Na+/H+ exchange. Shrinkage activation of both transporters follows a time lag of approximately 3 min and also requires normal levels of ATP. ATP depletion inhibits Na(+)-K(+)-2Cl- cotransport even at normal cell volume. Both transporters are partially inhibited by the protein kinase inhibitors staurosporine and K252a, and activators of protein kinases A and C do not affect transport. Okadaic acid inhibition of protein phosphatases activates Na(+)-K(+)-2Cl- cotransport to its maximal activity (same after shrinkage), but shrinkage and okadaic acid activation are not additive. In contrast, okadaic acid activates Na+/H+ exchange even in shrunken cells. These results indicate that cell shrinkage activates Na(+)-K(+)-2Cl- cotransport and Na+/H+ exchange probably by phosphorylation processes.

Occlusion of K+ in the Na+/K+/2Cl- cotransporter of Ehrlich ascites tumor cells

Proteins of n-octyl glucoside solubilized membrane vesicles derived from Ehrlich ascites tumor cells can occlude 86Rb+.K+ displaces 86Rb+ and it is assumed that 86Rb+ can be used as a tracer to measure K+ occlusion. The following observations indicate that the Na+/K+/2Cl- cotransporter is responsible for this occlusion: (1) Na+ does not compete for the K+ binding site, but rather stimulates 86Rb+ occlusion. (2) K+ occlusion saturates with increasing [Na+] and [K+], the respective K0.5 values being 50 +/- 7 microM for Na+ and 371 +/- 63 microM for K+. (3) Preincubation with 1 mM ouabain does not inhibit 86Rb+ occlusion, arguing against the Na+/K+-ATPase as being responsible for the occlusion. This notion is supported by the K0.5 value for K+ being higher than reported for Na+/K+-ATPase and by the stimulatory effect of Na+. (4) The K+ occlusion is sensitive to [Cl-], and the occluded ion is protected by the presence of bumetanide during cation exchange chromatography. Our results suggest that occlusion measurements of substrate ions could be a profitable way to study the ion binding mechanism(s) of the Na+/K+/2Cl- cotransporter.

The ATP and Mg2+ dependence of Na(+)-K(+)-2Cl- cotransport reflects a requirement for protein phosphorylation: studies using calyculin A

Na(+)-K(+)-2Cl- cotransport activity has previously been shown to depend on both intracellular ATP and Mg2+, but the mechanisms remain unknown. Cotransport in avian erythrocytes can be stimulated by a variety of agents including cAMP and permeant serine/threonine phosphatase inhibitors and is inhibited by prior depletion of either ATP with antimycin A, or mg2+ by incubation in A23187 plus EDTA. However, when cells were first stimulated with cAMP rather than calyculin A then subjected to either depletion strategy, a differential effect was found. The phosphatase-inhibitor-treated cells were resistant to subsequent ATP or Mg2+ depletion while cAMP-treated cells were sensitive to both treatments. Parallel examination of protein phosphorylation confirmed that ATP or Mg2+ depletion leads to dephosphorylation of membrane proteins in cAMP-treated but not in calyculin-A-treated cells. These results suggest that, for cotransport, ATP and Mg2+ are required primarily to maintain the system in a phosphorylated state rather than as direct modulators. The relative effectiveness of okadaic acid (EC50 approximately 630 nM) and calcyulin A (EC50 approximately 8 nM) in stimulating the cotransporter indicate that a type-1 protein phosphatase is probably responsible for dephosphorylating the system. Cells stimulated by hypertonicity were also resistant to ATP or Mg2+ depletion suggesting that the mechanism of shrinkage-induced cotransport stimulation might also involve protein phosphatase modulation.

Inhibition of Na-K-2Cl cotransport and bumetanide binding by ethacrynic acid, its analogues, and adducts

The inhibitory effect of ethacrynic acid (EA) and a variety of its derivatives on Na-K-2Cl cotransport in avian erythrocytes was investigated. The most potent compound tested was the adduct of EA with L-cysteine, with an IC50 of 7.2 x 10(-7) M. EA itself, dihydro-EA, EA-D-cysteine, and adducts of EA with other sulfhydryl (-SH) compounds were much less potent. The mechanism of action of EA and EA-L-cysteine differed in several respects: 1) EA-L-cysteine acted more rapidly than EA (half times of < 1 and 4 min, respectively, at 37 degrees C); 2) the action of EA-L-cysteine was reversible by washing, whereas that of EA was not; and 3) the degree of inhibition by EA-L-cysteine varied with medium [K], whereas that of EA did not. The inhibitory effects of both EA-L-cysteine and EA were affected by medium [Na] and [Cl]. We conclude that EA-L-cysteine does not "deliver" EA to transport-related -SH residues or act as an alkylating agent but has some stereospecific effect on cotransport that is a property of the entire molecule. EA does appear to inhibit cotransport by alkylating -SH residues, as closely related compounds lacking the ability to covalently react with such groups were reversible, and other -SH reagents (e.g., N-ethylmaleimide) also inhibited cotransport. EA, EA-L-cysteine, and EA-D-cysteine all inhibited [3H]bumetanide binding to membranes from activated avian erythrocytes at concentrations similar to those that inhibited cotransport. It is possible that the EA and bumetanide types of diuretics interact with closely apposed sites on the Na-K-2Cl cotransporter.

Regulation of vascular endothelial cell volume by Na-K-2Cl cotransport

The relationship between cell volume and Na-K-2Cl cotransport was studied in cultured bovine aortic endothelial cells. Hypertonic cell shrinkage increased bumetanide-sensitive, Na- or Cl-dependent K influx without altering bumetanide-insensitive influx. Greater stimulation of cotransport was observed in cells shrunken isosmotically either by preincubation in K-free and Na-free medium or by preincubation in hypotonic medium. Cell swelling, produced by preincubation in isotonic high-K medium, inhibited bumetanide-sensitive K influx. Simultaneous measurements of [3H]bumetanide binding and K influx revealed an increased number of binding sites without an increased influx per binding site in shrunken cells. Bumetanide did not alter the volume or ion content of cells in isotonic or hypertonic medium, indicating that no net influx of ions occurs through cotransport under these conditions. In isosmotically shrunken cells, there was greater stimulation of bumetanide-sensitive influx than of bumetanide-sensitive efflux, resulting in net bumetanide-sensitive influx. Rapid recovery of cell K, Na, and water occurred over 10-20 min and was inhibited by bumetanide or by the removal of external Na or Cl. These data demonstrate that Na-K-2Cl cotransport in aortic endothelial cells is regulated by cell volume, possibly through changes in the number of functional cotransporters, and mediates a brisk regulatory volume increase in isosmotically shrunken cells. Although thermodynamically favored, no net influx occurs through Na-K-2Cl cotransport in cells of normal volume or in hypertonically shrunken cells. This suggests additional regulation of cotransport, perhaps through trans-inhibition by intracellular Cl. Regulation of cell volume by Na-K-2Cl cotransport may be important in maintaining endothelial integrity.