During epithelial morphogenesis, the establishment of tight junctions precedes the development of both the asymmetry in protein and lipid composition between apical and basolateral cell surfaces (the 'fence' function) and the restriction in the transport of ions and nonelectrolytes through the extracellular clefts between cells (the 'gate' function). Molecular models that explain both functions envision strands of particles extending as rings in the cell's perimeter that interact with similar strands located at the apposing cell. This model accounts for the 'fence' function, because the strands prevent diffusion of protein and lipids, and also for the 'gate' function, because the interaction between strands minimizes the width of the extracellular clefts, increasing transepithelial resistance to ions and decreasing non-electrolyte permeability. Here we describe the results of energy depletion, which for the first time separates both functions: it abolishes the gate function, as determined by the dramatic decrease in transepithelial resistance, but it leaves the fence function intact, as determined by the maintenance of lipid polarity.

A linear relationship has been observed between the rate of active ion transport and the oxygen consumption or lactate production rate in a variety of epithelia. Stoichiometries of ions transported per oxygen consumed or ATP utilized calculated from these relationships reflect actual properties of the active transport step when the following two conditions are met: 1) the basal metabolic rate, obtained in the absence of active transport, remains constant at all rates of active transport; and 2) all the net transport across the tissue considered in the calculation traverses through active (energy-dissipative) pathways. The nature of the cellular mechanism linking active transport and energy production is a fundamental physiological question. Experimental alterations in the rate of active transport elicit mitochondrial state transitions and/or changes in adenine nucleotide concentrations in various epithelia. These findings, obtained by optical and biochemical methods, indicate that ATP and its hydrolysis products constitute part of the coupling mechanism linking the turnover of transport ATPase and the aerobic metabolic rate in epithelia.

An improved cortical tubule suspension from the rabbit kidney is described that contains almost entirely proximal convoluted tubules with little contamination by cellular debris or glomeruli. These tubules appear to be capable of active transepithelial transport, since their tubular lumina are open and ouabain inhibits 70% of the QO2. In previous preparations with closed tubular lumina, ouabain inhibited QO2 only 40% and the base-line QO2 was one-half to one-quarter that of the present preparation. The delivery of oxygen to this tubule suspension was compared to that of slices by three different means: oxygen consumption kinetics, ATP content, and, more directly by spectrophotometric monitoring of the redox state of cytochrome oxidase. Results demonstrate that the cortical slice is evidently oxygen deficient even when the bath PO2 is in excess of 570 mmHg. In contrast, the tubule suspension is shown to be adequately oxygenated even at a bath PO2 of 10 mmHg. This tubule suspension will be highly applicable for the analysis of aerobic metabolism in a functionally intact renal preparation by optical, electrode, and biochemical assay technologies.

Amiloride inhibited the ouabain-sensitive rate of oxygen consumption (QO2) of a suspension of rabbit intact proximal tubules in the presence of different concentrations of extracellular sodium. Measurements of the ouabain-sensitive QO2 in the presence of nystatin, the tissue sodium and potassium contents of the tubules in suspension, and the sodium- and potassium-dependent adenosinetriphosphatase (Na,K-ATPase) activity of lysed tubule membranes indicated that the effect of amiloride was due to a direct inhibition of the Na,K-ATPase activity of the proximal tubule.

Disruption of the renal proximal tubule (PT) brush border is a prominent early event during ischemic injury to the kidney. The molecular basis for this event is unknown. Within the brush border, ezrin may normally link the cytoskeleton to the cell plasma membrane. Anoxia causes ezrin to dissociate from the cytoskeleton and also causes many cell proteins to become dephosphorylated in renal PTs. This study examines the hypothesis that ezrin dephosphorylation accompanies and may mediate the anoxic disruption of the rabbit renal PT. During normoxia, 73 ± 3% of the cytoskeleton-associated (Triton-insoluble) ezrin was phosphorylated, but 88 + 6% of dissociated (Triton-soluble) ezrin was dephosphorylated. Phosphorylation was on serine/threonine residues, since ezrin was not detectable by an antibody against phosphotyrosine. After 60 min of anoxia, phosphorylation of total intracellular ezrin significantly decreased from 72 ± 2% to 21 + 9%6, and ezrin association with the cytoskeleton decreased from 91 + 2% to 58 + 2%. Calyculin A (1 ,uM), the serine/threonine phosphatase inhibitor, inhibited the dephosphorylation of ezrin during anoxia by 57% and also blocked the dissociation of ezrin from the cytoskeleton by 53%. Our results demonstrate that (i) the association of ezrin with the renal microvillar cytoskeleton is correlated with phosphorylation of ezrin serine/threonine residues and (ii) anoxia may cause disruption of the renal brush border by dephosphorylating ezrin and thereby dissociating the brush border membrane from the cytoskeleton.

The effects of various short-chain fatty acids, carboxylic acids, and amino acids on NADH fluorescence and oxygen consumption (QO2) of rabbit proximal tubule suspensions were determined. The short-chain fatty acids were the most effective substrates in increasing NADH fluorescence and QO2, followed by the carboxylic acids and amino acids. All of the substrates tested that increased NADH fluorescence proportionally increased QO2. This implies that the primary effect of these substrates was to increase QO2 by increasing the delivery of reducing equivalents to NAD and not by stimulating ATP hydrolysis directly. The relative affinity of several substrates to increase NADH fluorescence was also determined. The short-chain fatty acids had the highest affinity (10 microM range) followed by the carboxylic acids (100 microM range). These data demonstrate that the metabolic rate and NADH redox state of the renal cortical cell is very sensitive to the type of metabolic substrate available.

A B S T R A C TThe steady-state transport kinetics of the interaction between external sodium and the diuretic drug, amiloride, was studied in isolated anuran skin epithelia. We also investigated the effect of calcium on the amiloride-induced inhibition of short-circuit current (Ise) in these epithelial preparations. The major conclusions of this study are: (a) amiloride is a noncompetitive inhibitor of Na entry in bullfrog and grassfrog skin, but displays mixed inhibition in R. temporaria and the toad. A hypothesis which states that the interaction sites for amiloride and Na on the putative entry protein are spatially distinct in all of these species is proposed. (b) The stoichiometry of interaction between amiloride and the Na entry mechanism is not necessarily one-to-one. (c) The external Ca requirement for the inhibitory effect of amiloride is not absolute. Amiloride, at all concentrations, is equally effective in inhibiting Isc of bullfrog skin independently from the presence or absence of external Ca.

The association/dissociation of ezrin, a microvillar membrane-cytoskeleton linker, was studied to search for the initial step leading to anoxia-induced brush-border breakdown in a rabbit proximal tubule suspension. Electron microscopy studies display time-dependent damage to the microvilli during anoxia; immunoblots demonstrate the dissociation of ezrin from the cytoskeleton, reflected by the significant decrease in Triton X-100-insoluble ezrin from control (91%) to 39% after 30 min. Simultaneously, Triton X-100-soluble and extracellular ezrin increased with no change in total ezrin, Triton X-100 solubility of actin, or total intracellular protein. Parallel immunocytochemistry studies show diffusion of ezrin from the brush border, where ezrin is highly colocalized with F-actin during normoxia into the cytoplasm. Thirty minutes of reoxygenation following 30 min of anoxia causes recovery of the microvillar structure and reassociation of ezrin to the cytoskeleton and the brush border. Application of ethylene glycol-bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid (4 mM) or inhibition of intracellular calpain or calcineurin do not prevent the dissociation of ezrin during anoxia. We conclude that ezrin-cytoskeletal dissociation may initiate microvillar breakdown during anoxia via calcium-independent mechanisms.